Slow Electrical Processes in The Brain

P R O G R E S S IN B R A I N R E S E A R C H VOLUME 7 SLOW ELECTRICAL PROCESSES IN T H E B R A I N

PROGRESS IN BRAIN RESEARCH

ADVISORY BOARD

W. Bargmann E. De Robertis

J. C. Eccles J. D. French

H. Hyden J. Ariens Kappers

S. A. Sarkisov

Kiel Buenos Aires Canberra Los Angela Goteborg Amsterdam Moscow

J. P. Schadt

Amsterdam

T. Tokizane

Tokyo

H. Waelsch

New York

J. Z. Young

London

PROGRESS I N B R A I N R E S E A R C H VOLUME 7

SLOW ELECTRICAL PROCESSES IN THE BRAIN BY

N. A. ALADJALOVA Institute of Biophysics, Academy of Sciences, Moscow ( U . S . S .R . )

ELSEVIER PUBLISHING COMPANY AMSTERDAM

/

LONDON

1964

/

NEW Y O R K

ORIGINAL TITLE MEDLENNYE ELEKTRICHESKIE PROTSESSY V GOLOVNOM MOZGE P U B L I S H E D BY P U B L I S H I N G H O U S E O F T H E U.S.S.R. MOSCOW,

A C A D E M Y OF S C I E N C E S

1962

T R A N S L A T E D BY S C R I P T A T E C H N I C A , INC., W A S H I N G T O N , D.C.

(U.S.A.)

ELSEVIER PUBLISHING COMPANY

335

J A N V A N G A L E N S T R A A T , P.O.

BOX

211,

AMSTERDAM

AMERlCAN ELSEVIER P U B L I S H I N G COMPANY, INC.

52

VANDERBILT AVENUE, NEW YORK

17, N.Y.

ELSEVIER P U B L I S H I N G COMPANY LIMITED

128,

RIPPLESIDE COMMERClAL ESTATE

R I P P L E ROAD, B A R K I N G , ESSEX

L I B R A R Y OF CONGRESS CATALOG CARD NUMBER

WITH

103

ILLUSTRATIONS AND

64-15282

5 TABLES

ALL RIGHTS RESERVED T H I S B O O K O R A N Y P A R T T H E R E O F M A Y N O T BE R E P R O D U C E D I N A N Y F O R M , I N C L U D I N G PHOTOSTATIC OR MICROFILM FORM, W I T H O U T WRITTEN PERMISSION FROM THE PUBLISHERS

PRINTED IN THE NETHERLANDS

Other volumes in this series:

Volume I : Brain Mechanisnis Specific and Unspecific Mechanisrm of Sensory Motor Integration Edited b y G. Moruzzi, A. Fessard and H. H. Jasper

Volume 2: Nei.vr, Brain and Menlory Models Edited by Norbert Wiener and J. P. Schade

Volume 3 : The Rhinencephalon and Relared StrirctirreA Edited by W . Bargmann and J. P. Schade Volume 4: Growth arid Maturation o/’ the Brain Edited by D. P. Purpura and J. P. Schade

Volume 5 : Lectures on the Diencephalon Edited by W . Bargmann and J . P. Schade Volume 6 : Topics in Basic Neurology Edited by W . Bargmann and J. P. Schadk

Volume 8 : Biogenic Amines Edited b y Harold E. Himwich and Williamina A. Himwich

Volume 9: T/7e Developing Brain Edited by Williamina A. Himwich and Harold E. Him wich Volume 10: Structure and Frinc/ion of /he Epiphysis Cerebri Edited by J . Ariens Kappers and J. P. Schade Volume 11 : Organization of the Spinal Cord Edited by J. C. Eccles and J . P. Schade

Volume 12: Physiology of Spinal Neurons Edited by J . C. Eccles and J. P. Schade Volume 13 : Mechanisms of Neural Regeneration Edited by M . Singer and J. P. Schade

Volume 14: Degeneration Patterns in the Nervous System Edited by M . Singer and J. P. Schade

This Page Intentionally Left Blank

Contents Introduction

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

Chupter I . Recent data on the structure and.fiincrion of neurons in the cerebral cortex

Nerve elements and glia . . . . . . . . . Potentials on the neuron surface . . . . . Excitatory and inhibitory synapses . . . . Chemical transmitters . . . . . . . . . . Neuron excitability and dendritic potentials

1

. . . . .

5

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

5 10 11 13 16

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

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

Chnpter I I . D.c. potentials of the cerebral cortex

.................... cortex by their duration . . . . . . . . . . . .

Classification of potentials i n the cerebral Recording of slow potentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . Difference in potentials between different points and at various depths of the cerebral cortex Significance of extracellular currents in neuron function . . . . . . . . . . . . . . . Effect of external influences on the quasi-steady potential . . . . . . . . . . . . . . Changes in the quasi-steady potential in relation to the activity in dendrites . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter III . Infraslow rhythmic potential oscillations in the cerebral cortex

.

. . .

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

Infraslow potential oscillations in the hemispheres of a waking animal Factors that increase or decrease infraslow potential oscillations in the cerebral cortex . . Infraslow potential oscillations in the human brain . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

20 20 23 27 32 33 36 38 39 39

. .

45

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

60

Chapter I V . Excitability of neurons in the cerebral cortex and infraslow potential oscillations

Infraslow potential oscillations compared with the electrocorticogram Changes in cortical reactivity associated with infraslow potential oscillations . . . . . Infraslow potential oscillations in an isolated strip of cortex . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. .

. .

Chapter V . The role of certain subcortical structures in the appearance of infraslow potential oscillations in the brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

55 58

60 64 74 88

91

Infraslow potential changes in nuclei of subcortical and upper brain stem structures . . . . 91 Factors increasing infraslow potential oscillations in subcortical structures . . . . . . . . 99 Some examples of ISPO in the central gray matter and in the reticular formation in the brain stem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Acetylcholine and ISPO in the cerebral cortex and subcortical structures . . . . . . . . . 113 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Chapter VI . Slow rhythmic potential oscillations in the light of comparative physiological data

. 124

Slow rhythmic processes in certain elementary structures . . . . . . . . . . . . . . . .124 Living structures classified according to their manifestation of slow rhythmic processes . . . 129 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

. . . . . . . . 141 Change in the infraslow rhythm in formation of the conditioned defense reflex . . . . . . 141 Slow control system of the brain . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

Chapter VII. Infraslow processes in the brain as part of its integrative activity

CONTENTS

VlII

Chapter VIlI. Ionic processes in the cerebral cortex investigated by the method of conductivity

. 156

Method of investigating cortical impedance . . . . . . . . . . . . . . . . . . . . . Electrical parameters of the cerebral cortex in a waking animal . . . . . . . . . . . . . The state of excitation. inhibition and narcosis in relation to the electrical parameters of the cerebral cortex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dynamics of ionic changes in the cerebral ganglion of the crab . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

i56

i78 185 198 203

Chapter IX . Efect of ionizing radiation on the infraslow potential oscillations and electrical para207 meters of the brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary

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

213

References

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

215

Subject index

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

238

1

INTRODUCTION

Many phenomena in animal nerve tissue are accompanied by electrical processes. These processes are largely due to mechanisms of nervous activity. They provide special opportunities for studying those aspects of nervous activity which generally do not lend themselves to other research techniques. Conversion of the energy of cell metabolism to a specific neuronal function is responsible for molecular and ionic changes, which a r e associated with the appearance of electrical potentials and changes in electrical constants of the tissue - its electroconducting and dielectric properties. The electrical potentials that a r i s e in the brain have a varied origin. These include the brief impulses lasting 0.2-3 msec that follow excitation of the neuron and the slower potentials lasting 10-20msec that reflect the processes of local excitation in the cell body and in its dendritic processes. Slow potential oscillations with periods ranging from 50-500 msec are shown in the electrocorticogram (ECoG). Considerable information is now available on their origin and relationship with the electrical potentials that a r i s e in different segments of the neuron, but the subject is still moot in many respects. Electrical phenomena of an entirely different order and origin are also recorded in the brain structures, namely, infraslow rhythmic potential oscillations. By "infraslow'' rhythmic oscillations (Aladjalova, 1956a, b; 1957a, b) are meant the potential oscillations which are similar sinusoidal in form, with a period of 7-8 sec and an amplitude of 0.3-0.8 mV and with a period of 0.2-2 min and an amplitude of 0.5-1.5 mV; there a r e also intermediate values of the oscillation periods. Consequently, the oscillations range from a frequency of 7-8 to 0.5-2 per min. Electrical phenomena of the infraslow order are determined by processes connected with slow changes in excitable properties. They do not reflect direct shifts in response to excitation, although they are inseparably connected with them. The excitability of a neuron in the brain varies both with the impulses reaching it from other excited elements and with processes that are stimulated by humoral factors and the metabolic mechanism of nervous tissue. There are several types of interaction of nervous and humoral regulation. Certain humoral substances a r e released into the circulatory system under the influence of certain parts of the nervous system; circulating in the blood, they may influence the nervous system. In addition, excitation of an individual axon causes chemical substances to be released locally near the synapse. Consequently, the effect of humoral substances on nerve cells

2

INTRODUCTION

may be either localized o r generalized in character. This effect i s often selective with respect to the properties of individual components of the neuron. "We a r e unaware at present of any nervous regulation without the participation of humoral factors in some degree or form, nor a r e we aware of any humoral regulation that is not more or less connected with nervous regulation" (Orbeli, 1935). In the neurohumoral chain of phenomena, the nerve cell and the neuroglial cell must be regarded as a single functional unit. Intimate relations exist between these cells which a r e based on chemical processes and which a r e facilitated by the presence of a structural relationship resembling that between Schwann cells and nerve fibers. The humoral factors may effect the excitability of the neuron, acting through its dendrites, which possess high chemical sensitivity; the electrical manifestation of these processes may be very slow rhythmic changes in potential. Electrophysiology has at its disposal data on slow recordings of electrical potentials in the cerebral cortex of a steady characteristic (Kaufman, 1912; Libet and Gerard, 1941; Beritov; 1948b; Rusinov, 1953, 1958; Aleksanyan and Demirchbglyan, 1955; Leiio, 1944a, b, 1951; Bureg, 1954a, b, 1955, 1957; Goldring and O'Leary, 1951a, b, 1954, 1957; and others). However, in most of the familiar observations these changes are aperiodic, being directly related in time to a stimulating effect on the nervous tissue. Our main task was to investigate the phenomenon of the infraslow rhythmic potential oscillations that are found in various brain structures of warm-blooded animals, the origin of which could not be directly linked to concurrent nervous excitation. Infraslow potential oscillations a r e manifested differently in the various brain structures, which may be divided into two categories accordingly. In the rabbit, infraslow rhythmic potential oscillations may a r i s e in the cerebral cortex and hypothalamus, chiefly in the dorsomedial nucleus and premammilary region. Infraslow potential oscillations are not found in the nuclei of the thalamus, in the central gray matter around the walls of the aqueduct of Sylvius, o r in the brain stem reticular formation. Slow oscillations that take a long time to become extinguished are characterized by the manifestation of vital activity at different stages in evoluti-n. The existence of these oscillations is ensured by the fact that the actual process is controlled by the inflow of energy required to maintain it, and for this reason we call it an auto-oscillary process. Auto-oscillations are also encountered in the simplest forms of life. They a r e even found at the level of plasmodium protoplasm (Kamiya and Abe, 1950) and in some muscle structures (Koshtoyants, 1957). Auto-oscillatory processes appear in the form of a cyclic change in mechanical properties in some structures and in the form of physicochemical changes in others (Aladjalova, 1950b; Aladjalova and Mertsalova, 1954). The frequency of oscillations is influenced by physicochemical and chemical factors, e .g. by epinephrine (Aladjalova, 1950b) and acetylcholine (Bulbring, 1957) in smooth-muscle structures,

INTRODUCTION

3

by acetylcholine in ciliary movement (Bulbring et al., 1953), and by serotonin in rotation of mollusc embryos (Manukhin and Buznikov, 1960). Hormones influence the parameters of certain auto-oscillatory processes in the tissues of higher animals. This is also shown by an increase of the infraslow potential oscillations in the rabbit brain after activation of the animal's hormonal functions. Living structures may be divided into those in which very slow oscillations of electrical potential appear in the course of normal activity and those in which this process arises only under certain conditions (chronic denervation, exposure to chemical agents). The first type includes several types of smooth muscles and some types of nerve cells in the ganglia of invertebrates. The second type includes the skeletal muscles, nerve fibers, and certain types of nerve cells. It is interesting to note that the magnitude of the infraslow rhythmic process in central nervous system structures is largely proportional to the sensitivity of these structures to certain chemical factors in the environment Infraslow cerebral activity is probably intensified by the occurrence of certain chemical gradients in nervous tissue and in this respect r e flects the activity of the slow controlling system that ensures control of the level of activity in relation to the operation of the mechanisms that maintain stability and homeostasis. Much slower periodic processes a r e also to be found in the brain. Specifically, cycles of activity a r i s e after external perturbation in seve r a l structures with 20-30 min periods and last for several hours ("hour -long os:illations"). Assuming that the control systems of the brain a r e subdivided into high- and low-speed infraslow activity and "hour -long oscillations" should be regarded as a manifestation of slow system control. Since the phenomenon of infraslow rhythmic potential oscillations was not previously investigated, we thought it worthwhile to subject it to a detailed analysis using a variety of experimental techniques. It w a s necessary to find answers f o r the following questions. What brain structures a r e characterized by the phenomenon of infraslow rhythmic changes in potential? How are they related to the overall excitability of nervous t i s s u e ? Are all kind; of neurons equally involved in this phenomenon? What influences effect the frequency and amplitude of infraslow potential oscillations in the brain? What is the role of neuroendocrinal relations in this phenomenon ? It was important to obtain some idea of the possible mechanisms governing the origin and regulation of the infraslow rhythm and of the nature of the processes reflected in the infraslow potential oscillations. If these processes are of significance for the integrative activity of the brain, they should change regularly as new temporary connections a r e formed. It was therefore important to investigate this phenomenon in the human brain. Neurohumoral interactions in the cerebral cortex a r e reflected not only in the phenomenon of infraslow rhythmic potential oscillations, but also in the physicochemical properties of nervous tissue which determine its electroconductivity. Measurements of the electrical

.

4

INTRODUCTION

impedance may help to elucidate the mechanism of these ionic shifts. Impedance of the cerebral cortex can be investigated in uivo in wakeful animals by means of implanted electrodes (Aladjalova, 1953a, 1954a, 1955a). Two factors are responsible for a change in cortical impedance. One characterizes the general level of ionic mobility in nervous tissue and changes more or less simultaneously with a change in the metabolic level of the cortex and physiological condition of the animal. The other characterizes the local ionic shifts in the upper layers of the cortex and undergoes slow oscillatory changes. These changes are closely related to the nature of the electrical activity of the cortex and seem to occur largely (depending on the arrangement of the electrodes) in the layer of the apical dendrites, A change in the physicochemical conditions of the dendritic a r e a precedes a change in excitability of the neuron bodies. This raises the question of the part played by the exchange between dendrites and the environment in regulating the metabolic level of the neurons. In studying the mechanism of the biological action of physical agents, drugs, etc., the various approaches that have been suggested seem promising. For example, investigation of the slow electrical processes in the brain following exposure to ionizing radiation made it possible to distinguish two types of cortical neurons by their response to irradiation, to identify the dendrites as the most reactive element of the neuron, and to broaden the analysis of the origin of the phases of postradiation changes (Aladjalova, 1957). Since it has been shown that infraslow potential oscillations reflect certain chemical processes in nervous tissue, specifically those connected with neurohormonal relations and with the function of the neuroglia, recording of these oscillations may add new information to our knowledge of the function of the brain. In view of the fact that changes in neuroglial activity play a part in the pathology of brain metabolism, it is fair to assume that the proposed method will find application in medical practice.

5 CHAPTER I

RECENTDATAONTHESTRUCTURE ANDFUNCTIONOFNEURONS

IN THE CEREBRAL CORTEX

The structure and function of nerve elements of the cerebral cortex a r e being investigated at three levels: (1) cellular constituents nucleus, cytoplasm, and surface layer; (2) the neuron as a whole and its connections with other neurons; (3) nerve complexes linked togethe r by influences that often encompass whole divisions of the nervous system. We shall briefly describe recent knowledge of this field, emphasizing those aspects that are pertinent to our investigation. NERVE E L E M E N T S AND GLIA

Sarkisov and Polyakov (1949) distinguished various types of neurons in the cerebral cortex. PyrunziduZ tzeuyons have a cell body from 10-40 p or more in diameter and two types of dendritic branches: apical dendrites several millimetres long extending upward to the pia mater and basal dendrites with branches near the base of the cell body in an a r e a with a radius of 150-200 p (Fig.1). One neuron can have more than 50 dendritic branches. Pyramidal neurons may be further subdivided according to the course of their axons. The axon may emerge from the center of a cell body and in certain neurons travel through the white substance of the hemispheres to other parts of the nervous system. In other neurons it turns toward the surface of the cerebral cortex; o r it ramifies and its processes return to the region of their own basal dendrites where they make synaptic contacts (recurrent collaterals). Stellute neurons with dendrites ramifying near the neuron body. The axons an3 dendrites branch out in all directions and sometimes extend into neighboring a r e a s of the cortex. Many axons coming from the white substance terminate around these stellate cells. Other types of neurons, spindle shaped, spider, etc.? a r e varieties of these basic cells. Distribution of the neurons in the cortex varies from layer to layer. The deeper one proceeds in the inner third of the cerebral cortex, the more numerous the larger neurons are. The number of stellate cells decreases in the deeper layers while the number of pyramidal cells increases. Finally, the stellate cells with their locally distributed axons a r e again common among the pyramidal cells in the bottom layer. The density of the neurons in the cerebral cortex varies from layer

6

STRUCTURE AND FUNCTION O F NEURONS

B

A

Fig. 1. (A) Pyramidal neuron from the sensorimotor cortex of the cat: 1 = apical dendrite: 2 = basal dendrites; 3 = cell body; 4 = axon; 5 = recurrent collateral of axon; (B) oligodendroglia immediately adjacent to the neuron (satellite cells) (Penfield, 1924): (C) neuroglial cell stained by Golgi's method (Glees, 1955).

to layer. Figures for the cat visual cortex by Sholl (1956) are presented in Table I. Due to the neuron density in the cerebral cortex and the abundance of axonal connections between them, the activity of a single neuron may TABLE

I

DENSITY OF NEURONS IN THE CAT VISUAL CORTEX (SHOLL, 1956)

Depth (iii p fvoni Depth (in p f r o m No. of cells the piu m t t e r ) p e r 0.001 n1nz3 the pia matev)

N o . of cells per 0.001 rnm ~

0-150 25 0 350 550 650 75 0 850

106 97 43 40 66 69

1150 1250 1350 1550 1650 1750 1850

46 83 97 46 57 26 80

NERVE ELEMENTS AND GLIA

7

effect the excitability of 4000 other neurons through the synaptic contacts. The s o n s of the associative neurons that e m e r g e from different r e gions of the cortex proceed t o other regions and terminate in all l a y e r s except the most superficial one (Sarkisov, 1956). The specific afferent fibers, which are p a r t of the classical sensory pathway from the r e c e p t o r s through the specific nuclei of the thalamus, proceed through the lowest layer of the cortex and enter the central p a r t of the cortex in the region of the stellate cells. F o r example, afferent fibers from the lateral geniculate body may ramify into 12 branches (the distance between the furthest branches is about 650 p) and end at different depths in the region of the stellate cells of the visual cortex. Along the pathway of such branches 5000 neurons may be encountered. The excitation threshold of these branches is often changed by afferent impulses. Some 25,000 afferent and 75,000 efferent fibers p a s s through a 1 mm" c r o s s section of visual radiation in a cat (Sholl, 1956). The distance between the c e n t e r s of the f i b e r s is no m o r e (Bok, 1959). than 0.4-1.9 Such density of conduction pathways r e q u i r e s them to be fairly well insulated from one another. Otherwise, the excitation impulses arising in some conductors could change the excitability of the adjacent fibers by influencing the electrical field. The neuroglia may partly function as insulation. Numerous axonal endings approach the dendrites from other neurons. The receiving surface of the cell body is known t o constitute only 10% of the receiving surface of the dendrites (Sholl, 1956). The decrease in density of the basal dendrites with distance from the cell body limits the receiving area of the neurons in the region of the basal dendrites t o a radius of about 100 u. Most of the apical dendrites enter layer I, which contains both myelinated and small unmyelinated fibers that proceed a considerable distance parallel to the cortical surface. These fibers a r e the axons of different neurons in the subjacent layers. The dendrites differ f r o m the axons in physical and chemical properties as well as in mechanism of excitation, A major problem in the histology of the c e r e b r a l cortex is t o elucidate the organization of the synaptic endings. Unlike the motor neurons of the spinal cord, the cortical c e l l s have not yet been definitely proved to have special axonal endings on the cell body. Cases have been observed in which the axons s e e m to entwine the cell body. Axons are frequently parallel and very close to dendrites, It is also known that the apical and basal dendrites are interspersed with gemmules which, according to Sarkisov and Polyakov (1949), may function as synaptic contacts. On the basis of anatomical and physiological data, Chang (1951a, b) divided synaptic contacts in the c e r e b r a l cortex into two categories: pericorpuscular and paradendritic. The f o r m e r a r e formed on the cell body, the latter on the dendrite. The high density of the pericorpuscular synapses makes it very probable that postsynaptic potentials develop in the cell body. The number of axons approaching the cells and their methods of contact

STRUCTURE AND FUNCTION OF NEURONS 8 with the cell bodies and dendrites have a bearing on the distribution of information through the cortical neurons. They integrate the information coming from different sources and control the excitability of the cell. Palade and Palay (1954, 1955) detected some details of the fine structure of synapses of different cells by means of investigation with the electron microscope. Within the synaptic ending could be seen vesicles 300 A in diameter which were at times in direct contact with the membrane, while at other times they were adjacent to the mitochondria. The presence of mitochondria in a synaptic ending is evidence of high metabolic activity therein. It is believed that the vesicles contain a chemical substance (acetylcholine) that takes part in the transmission of excitation through the synapse (Robertson, 1956). Neurons interact as a result of chemical processes in the synapse and neuroglia. There are two main types of neuroglia: macroglia and microglia. The former consists of protoplasmic and fibrous astrocytes (Fig. 1)with a large nucleus and many branching processes as well as oligodendroglia cells (with a small nucleus), which were once thought to have no branches until Penfield (1930) demonstrated the opposite to be true. The microglia consists of long branching cells. The astrocytes are about 10 p in diameter. Their processes approach and parallel the blood vessels for a distance of 5-7 u. The astrocytes come in contact with the neuron and oligodendrocytes through their heavily branched network. The oligodendroglia is located very close to the nerve cells (Fig. 1). It is indirectly connected with the blood vessels through the astrocytes. Certain cells of the oligodendroglia are approached by a few axon collaterals, which either terminate at the body of the oligodendrocytes o r touch them and then travel on (M. and A. Scheibel, 1958). Multiple contacts exist between the oligodendroglia and dendrites. In the rabbit (Glees, 1955), fibrous astrocytes with short branches can be easily seen in the upper layers of the cerebral cortex; their ramifications terminate on blood vessels or on the pia mater. Heavily branched fibrous astrocytes with a large nucleus are frequently found in the deep layers of the cortex. In layers II and IXI glial cells lie very close to one another. Protoplasmic astrocytes occur in layers IV and V. Their number decreases from the surface of the cortex to the white substance, while the number of oligodendroglia cells increases. Oligodendrocytes in the white substance acquire long branches, which proceed along the nerve fiber s . Neuroglial elements are more strongly developed in man and monkey than in animals standing lower on the evolutionary ladder (Achucarro, 1918). Some authors assumed that the ratio of the number of glial cells in the cerebral cortex to the number of nerve cells indicates the stage of brain development. However, this view was refuted by the discovery that this ratio is higher in whale than in man. The ratio of glial cells to nerve cells most likely increases with the s i z e of the brain because the neuron density decreases at the same time: in mice: 142.5 neurons per 0.001 mm3 of cerebral cortex; in rabbits: 43.8; in dogs: 24; in men: 10.5; in whales and elephants: 6.8 (Tower, 1954).

NERVE ELEMENTS A N D GLIA

9

The close apposition of the glia to the blood vessels makes it likely that the glia constitutes a unique b a r r i e r between nerve elements and the blood system, The surface of the cortex has a layer of glia in the pial region that separates nervous tissue from the fluid compartment in the subarachnoidal space (Zavarzin and Schelkunov, 1954). It has been conjectured that the distance between the cell body and synaptic endings of the incoming axons may change as the glia swells and thus influence conduction in the synapse. However, Palay and Palade (1954) concluded, on the basis of an electron microscope study, that there is no neuroglia between the synaptic endings and the cell body. The role of the neuroglia in synaptic transmission may be manifested through i t s effect on neuron metabolism (Snesarev, 1926). The neuroglia has an active metabolism, the rate in the oligodendroglia being higher than in the other forms of glia. The neuroglia may supply the neuron with nutrients synthesized by its metabolism. This is probably the function of the satellite cells. The neuroglia can therefore b e regarded as a mechanism that ensures the metabolic level needed to excite the neuron. This mechanism may well be working through the axon-oligodendrocyte synaptic system (M. and A. Scheibel, 1958). Hence the inflow of information through the collaterals of the axons to the oligodendroglia may serve to prepare the neuron for activity. According to a hypothesis of Galambos (1961), oligodendroglial cells may become storehouses of information, which is imprinted chemically as a result of signals from the axons. The earlier theory that the neuroglia provides mechanical support for the nerve elements was recently contradicted by new findings. It was found that the position of the neuroglia i s not permanently fixed (Horstmann, 1954) and that its processes change position, increasing o r decreasing their sphere of influence. Moreover, the cell body of glia (oligodendroglia) pulsates rhythmically in a tissue culture (Lumsden and Pomerat, 1951). The glial cell may contract in response to electrical stimulation (Chang and Hild, 1959) with a latent period of 1.5-4 min; length of contraction: 1.4-3.4 min, relaxation: 6-16 min. Glial cells possess a resting membrane potential of the order of 50 mV recorded with an intracellular electrode (Hild et a l . , 1958). The glial cell membrane is depolarized in response to strong electrical of it's magnitude* for 4 sec, e i.e. 1000 times longer than in neuron excitation. The neuroglia in the ganglia of some invertebrates contains protein granules (Scharrer, 1941), a neurosecretory substance** liberated very close to the nerve cells and participating in their metabolism.

stimulation, the potential decreasing by

* e - the base of natural logarithms (e z 2.7) se:ves as the characteristic to determine the time constant. ** Neurosecretion has now been found in many nerve structures. F o r example, spherical inclusions of different s i z e s have been noted in the nerve cells of the earthworm nerve cord (De Robertis and Bennett, 1955) and they participate in synthesis of the secretion (presumably epinephrine). Neurosecretory cells have a l s o been described in several subcortical nuclei (hypothalamus, amygdaloid nucleus, etc.) in vertebrates (Barry, 1954a, b, c; 1956).

10

STRUCTURE AND FUNCTION O F NEURONS

Secretory material 1 . also present in neuron cytoplasm in the form of granules 400-2000 A in diameter. The cytoplasm has tiny vacuoles (0.5-1 p in diameter) filled with homogeneous material and solid masses 200 thick lying alongside. The structure of neuron protoplasm includes molecules of nucleic acids arranged in such a way that desoxyribonucleic acid is concentrated in the nucleus and ribonucleic acid in the cytoplasm (Hyden, 1943). Differentiation of cytoplasm into a system of membranes of different kinds facilitates the organization of enzymatic reactions and i s a controlling factor in cell metabolism. POTENTIALS ON THE NEURON SURFACE

Measurement of the electrical resistance of cytoplasm has shown that, despite the structural heterogeneity of the latter, it is very low compared with the resistance of the surface membrane of the cell. Consequently, most authors regard cytoplasm as an isopotential system. Yet there i s evidence of a potential drop of about several millivolts between different p a r t s of the cytoplasm in neurons. This drop was measured in a giant nerve cell of the mollusc Aplysia (Arvanitaki and Chalazonitis, 1956). This potential difference is of a lesser magnitude than on the surface membrane on which it reaches 70 mV in a resting state (negative on the inside). It is assumed that this potential difference on the surface membrane of the cell body is virtually the same all over the membrane. However, points on the surface membrane far away from the cell body, e . g . on the smaller dendrites, will nevertheless have different potential values because the diameter of the dendrite causes it to have a high electrical resistance (Coombs et a l . , 1955a, b). Thus a shift in potential on the cell body when it is excited spreads only to the segments of the dendrites nearest to the body, thereby embracing an area with a radius of about 70-100 u. Some difference in potentials a r i s e s between this a r e a and an a r e a with a diameter of 300 I-I to which the dendrite branches extend. The potential generated by the cell body upon excitation apparently does not reach the branches of the apical dendrites, but it may activate the basal dendrites (Clare and Bishop, 1955a). The potential generated by the apical dendrites is caused by the excitation reaching the branches of the synaptic endings. The resultant postsynaptic potentials a r e summed and a dendrite potential is created. Dendrite potentials do not generate an impulse and they spread only electrotonically (Clare and Bishop, 1955a, b). The membrane potential of the cell body is measured with a microelectrode 0.2-0.5 ,iin diameter impaled in the cell, with an indifferent electrode at a remote point. The membrane potential is 60 mV for a pyramidal cell of the cat cerebral cortex (Phillips, 1955, 1956a, b), whereas it is somewhat larger (70 mV) for a spinal cord motor neuron in the same animal (Frank &nd Fuortes, 1955). The lowest membrane potential has been recorded in the ganglion cell of the snail (from 30 to 60 mV) (Tauc, 1954). The membrane potential of the axon in this

EXCITATORY AND INHIBITORY SYNAPSES

11

animal is equal to the cell body potential of the same neuron (Eyzaguirre and Kuffler, 1955). Regarding the motor neuron (Eccles, 1957), a high-voltage action potential (80-100 mV) is generated in the cell body membrane and adjacent portions of the dendrites by a complex s e r i e s of processes. The threshold of this membrane is much higher than the threshold of the membrane of the axon hillock or the so-called "initial segment" of the axon. It has been shown that a spike potential a r i s e s first in this segment, and a similar potential being generated on the somadendritic membrane later on. EXCITATCItY AND INHIBITORY SYNAPSES

Neuron excitation and inhibition are caused by synaptic processes that may evoke or facilitate the formation of an action potential or, contrariwise, prevent it from developing. A presynaptic stimulus results in the formation of a local postsynaptic potential. An action (spike) potential a r i s e s as a result of a triggering impulse, i.e. a depolarization potential that has reached a critical magnitude. On the other hand, comparatively slight hyperpolarization (postsynaptic hyperpolarization) blocks an action potential. The magnitude of the action potential of the cat cortical pyramidal cells i s 80 mV (Phillips, 1956a, b) while its duration is 1 msec. It has been conjectured that depolarization of the postsynaptic membrane requires the formation of a current that enters the cell body near the synapse. It has been calculated for the cat motor neuron that such an incoming current caused by electrical processes in the presynaptic terminals increases to a maximum in 0.5 msec and then decreases after 1.2 msec (Coombs et al., 1956). At the same time the postsynaptic potential reaches a maximum only after 1 msec, 2.e. when the incoming current of presynaptic origin has become virtually imperceptible, and then slowly decreases after 10 msec or more. These facts can be explained not by the hypothesis of electrical transmission in a synapse, but by the hypothesis of chemical transmission. The electrical transmission hypothesis likewise fails to account for the hyperpolarization of the postsynaptic membrane which is caused by inhibiting impulses (Brock et al.,1952). There is as yet no confirmation of Vorontsov's assumption that the inhibiting synapse has a special structure that permits hyperpolarization of the postsynaptic membrane by the mechanism of electrical transmission. An inhibitory stimulus increases the permeability of the cell body membrane for potassium and chloride ions, which causes hyperpolarization of the entire membrane. This increase in ion permeability r e duces the effectiveness of the excitatory stimulus. A relatively long inhibitory postsynaptic potential creates the conditions for the summation of inhibitory influences. The effectiveness of this mechanism also depends on the antagonistic process of depolarization created by exciting stimuli. A neuron creates an action potential only when the effect of exciting stimuli is dominant at some particular

12

STRUCTURE AND FUNCTION OF NEURONS

moment and causes depolarization above a critical level. This critical level is 10 mV €or the initial segment of the axon and about 20 mV for the soma and dendrites (Eyzaguirre and Kuffler, 1955). These values were also determined in e q e r i m e n t s in which depolarization w a s caused by a current from an artificial source (Araki and Otani, 1955). The application of a direct current through microelectrodes (intracellular and extracellular) causes the potential to change simultaneously on the cell body membrane, in the proximal region of the dendrites, and in the initial segment of the axon (Eccles, 1957). The remote r e gions of the dendrites are less exposed to the effect of this current. The greater the strength of the depolarizing current, the lower the threshold and the shorter the.latent period of generation of an action potential in response to natural o r artificial stimulation. Meanwhile the amplitude of the exciting postsynaptic potential decreases. A hyperpolarizing cur rent causes changes of the opposite kind. Thus, the application of current from a n external source may cause the same changes in membrane potential as natural stimulation. Hyperpolarizing postsynaptic potentials were also detected by extracellular recording of cortical cells (Phillips, 1956a and b) and by means of pharmacological analysis of cortical dendrites (Purpura and Grundfest 1957). The existence of hyperpolarizing postsynaptic potentials in the dendrites is usually masked by the presence of depolarizing potentials. A selective blockade of the latter by y-aminobutyric acid unmasks the hyperpolarization. Purpura and Grundfest (1957) think that inhibitory and excitatory potentials exist separately in the cerebral cortex. A primary blockade of the inhibitory synapses creates the conditions for the appearance of excitation. For example, the exciting effect of strychnine and moderate doses of d-tubocurarine is the result of a blockade of the inhibiting synapses. According to Grundfest (1957a), the cell membrane apparently has two mechanisms. One is excited only by chemical means, and it reacts by producing local postsynaptic potentials. The local currents that arise at this time activate, in turn, the other mechanism, one that has already been excited electrically, and which generates a spreading action potential. The basis for the assumption that nerve cells possess a heterogeneous membrane, some parts of which yield a postsynaptic potential while others yield a spike potential, was determined experimentally in the lobster cardiac ganglion, where a spike potential arises at some distance from the point of origin of the postsynaptic potential (Hagiwara and Bullock, 1957). There is some indication that pharmacological agents can be used to block the excitation of a chemically excitable postsynaptic membrane, whereas an electrically excitable membrane retains its capacity to generate an action potential. After employing this technique, Grundfest (1956, 1957b) concluded that dendrite potentials have a chemical postsynaptic mechanism and are not excitable electrically. Dendrites differ from axons in many respects. Dendritic potentials do not follow the "all or none" law; they spread decremental; they are capable of summation; they lack refractoriness. Dendrites are not

CHEMICAL TKANSMITTERS

13

excitable electrically, but they have a high chemical sensitivity. Axons, on the other hand, are excitable electrically; they have an absolute and relative refractory period; they are not capable of summation; they conduct excitation without decrement and at a certain velocity; they obey khe law of "all or none"; they have a short latent period and a r e weakly sensitive to chemical agents. Dendritic potentials a r e a less specialized form of excitation than the axonal potentials and are even found in primitive nervous systems. The electrotonic character of the dendrite potentials is a more general phenomenon than the action potential, which a r i s e s at a higher level of development (Clare and Bishop, 1955a). Thus, only a postsynaptic potential is produced in the apical dendrites by means of the synaptic mechanism, but no impulse is formed in the dendrites. It w a s assumed at one time that the postsynaptic potential in the neuron can be evoked both by chemical and by electrical stimulation. It was later shown that some types of peripheral synapses are not excitable electrically; the electrical organ of several fishes (Grundfest, 1957a), the end plates of muscle (Castillo and Katz, 1956), and the smooth musculature of frogs (Kuffler and Vaughan-Williams, 1953). Electrical nonexcitability of the postsynaptic membrane was found in the apical dendrites of cortical neurons and in spinal cord motor neurons (Eccles et al., 1954). Many investigators a r e inclined to believe that electrical nonexcitability of the postsynaptic apparatus of the nerve cells and transmission of excitation from the axon to the nerve cell through a chemical transmitter released in the synapse, is a universal phenomenon. CHEMICAL TRANSMITTERS

Samoilov (1925) was the first to formulate the chemical principle of nervous transmission in its general form. He wrote: "Wherever the process of excitation is to be transmitted from one cell to another, be it Sherrington's synapse or be it the boundary between the efferent nerves and effector, we shall find the characteristics of excitatory transmission - loss of time, one-way transmission, summation, etc. we assume that of two continuous cells, one has developed the capacity to liberate a stimulating substance, while the other has developed the capacity to react to this substance". Data on the chemical transmission of excitation by means of acetylcholine were first obtained in the neuromuscular synapse (Dale, 1935) and sympathetic ganglion (Kibyakov, 1933; Shevelyova, 1945, 1953). According to P e r r y ' s hypothesis (1953), free acetylcholine, which i s rapidly inactivated by cholinesterase , is biologically active, Cholinergic type transmitters are also present in synapses of the central nervous system: in the spinal cord, cortex, and subcortical structures of the brain (Bulbring and Burn, 1941; Feldberg, 1945; Mikhelson, 1948; Feldberg and Vogt, 19481. Feldberg (1954a) and Machtosh and Oborin (1953) found that a great deal of acetylcholine (up to 6 x lo4 g/cm/min) i s released from the cerebral cortex while it

14

STRUCTURE AND FUNCTION OF NEURONS

is active and that this release is absent during anesthesia. However, it must be borne in mind that tissue (glial) acetylcholine, a humoral factor of nonsynaptic origin, plays an important role in the cerebral cortex (Desmedt and La Grutta, 1955). Electron microscopic data show that the membrane of nerve endings have vesicles about 300 %, in diameter which may appear to the surface (Robertson, 1956), causing, as in the case of the neuromuscular end plates, miniature potentials to develop. Prolonged depolarization of the membrane may also be caused by the escape of these vesicles, which apparently contain the transmittor substance. The concentration of vesicles in the presynaptic endings is much higher than in the nerve trunk, and it corresponds to the analogous distribution of acetylcholine concentrations (MacIntosh, cited by Eccles, 1957). A s the rhythmic presynaptic impulses flow in, the vesicles are increasingly mobilized, a fact that may be related to the phenomenon of post-tetanic facilitation. The action time of the transmitting substance is several fractions of a second, but it may be lengthened by inhibition of cholinesterase; it exceeds the time in which the substance could be diffused into the s u r rounding medium. This fact led Eccles (1959) to posit the existence of a b a r r i e r around the synapse that prevents the diffusion of acetylcholine from under the synapse. Cholinesterase can also be conceived of as being included within the barrier. Such a b a r r i e r could account for the prolonged action of the transmitting substance and the phenomenon of subsequent facilitation. It could likewise explain the occasional ineffectiveness of the method of applying or intra-arterially injecting substances. However, not all the synapses of the cerebral neurons are cholinergic (Feldberg, 1954a, b). There are other pharmacological agents that excite central nervous system neurons, e .g. adenosinetriphosphate and substance P, a polypeptide (Von Euler and Gaddum, 1931). Factor P is found in the gray substance and in certain spinal cord systems (Pernow, 1953). It is present in large concentrations in mesencephalic nuclei, the hypothalamus, and corpus striaturn; it is present in low concantrations in the cerebral cortex,. optic nerve, and cerebellum. Serotonin, or 5-hydroxytryptamine (Twarog and Page, 1953), is another active substance that is similarly distributed in the brain. Experiments on smooth muscles (Biilbring and Lin, 1957) have shown that serotonin may act as a nerve mediator. Factors that influence serotonin metabolism have a marked effect on central nervous system activity. Serotonin by itself has a tranquilizing effect on animal behavior. A similar effect of reserpine is probably based on the rapid mobilization of brain serotonin. F r e e serotonin is inactivated by monoaminooxydase. Blocking of the latter by iproniazide promotes the accumulation of free serotonin. A well-known neurohumor is norepinephrine, a mediator of postganglionic transmission of excitation in the sympathetic nervous system. The significance of norepinephrine for central nervous system synapses has not yet been finally determined, although it is apparently related to nervous processes in several divisions of the central nervous system, e.g. in the hypothalamus, where the amount of norepinephrine

CHEMICAL TRANSMITTERS

15

diminishes after prolonged activity (Vogt, 1959). Recent knowledge of central synaptic transmitters is still inadequate and, except for acetylcholine, no other substance (epineprhine, nor epinephrine, serotonin, substance P, histamine) has been demonstrated to act as a central synaptic transmitter. Factor I, isolated from the cerebral cortex, possesses the properties of an inhibitory transmitter (Florey and McLennan, 1955). They conjectured that its active component is y-aminobutyric acid, which is present in large quantities in the central nervous system. McLennan (1957) later found that another substance with a more complex structure is the active component. The inhibitory effect of y-aminobutyric acid i s manifested particul a r l y by the fact that when applied to the cerebral cortex, it causes an alteration in the sign of the superficial response of the cortex due to action on the axodendritic synapses superficially located in the cerebral cortex (Purpuraet al., 1957b). Its inhibitory effect is due to a blockade of the depolarization processes (2.e. a blockade of the excitatory synapses). Investigation of various amino acid derivatives of this type has shown that some of them a r e able to block the depolarization effect and thereby induce inhibition while others block the inhibitory synapses, thereby creating excitation. Grundfest (1957), analyzing the molecular configuration of transmitting substances, assumes that the blockade of synaptic conduction is caused by the length of the carbon chains and by the polarity of the carboxyl and amino groups of different o-amino acids (Purpuraet a l . , 1957a, b). After comparing the effects of y-aminobutyric acid and strychnine, a strong central stimulator, Purpura et a l . (1957b) concluded that strychnine blocks the inhibitory synapses, whereas y -aminobutyric acid blocks the excitatory synapses. Elliott (1959) adopted a different view, arguing that y -aminobutyric acid acts selectively on the dendrites superficially located in the cerebral cortex, whereas strychnine increases the excitability of the synapses in the deeper layers. The mechanism of action of strychnine as an agent that blocks the inhibitory synapses in a more direct fashion was demonstrated in the motor neurons (Bradley et al., 1953). According to one of the postulates of modern nemopharmacology, all the axonal branches of a given nerve cell release the same mediator when excited (inhibitory or excitatory). Accordingly, one class of nerve cells will function exclusively as inhibitors, another as excitors (through all their synapses), i.e. two types of cells can be distinguished in t e r m s of function. Renshaw cetls in the spinal cord are an example of exclusively inhibitory neurons. The motor neurons and some of the internuncial neurons of the spinal cord a r e only excitatory. According to Eccles (1957) inhibitory neurons a r e mainly those neurons with short axons entirely localized in the gray substance. The neurons which give off their axons into the conduction pathways are generally excitatory. This principle of differentiating the neurons should also be taken into account with respect to the cortical neurons. However, Grundfest (1957) thinks that the same transmitter, released in a synapse, will produce a different effect depending on the nature of the postsynaptic membrane.

16

STRUCTURE AND FUNCTION OF NEURONS NEURON EXCITABILITY AND DE NDRITIC POTENTIALS

Apical dendrites of the pyramidal cells lying parallel to one another are convenient objects for investigating the nature of the dendrite processes. One of the methods of doing so is to study the effects of direct stimulation of the surface of the cerebral cortex. These effects a r e manifested in the development of an electrical response, the negative potential of which is dendritic in origin ( e . g .Adrian, 1936; Beritov, 1948b; Roitbak, 1953a; Clare and Bishop, 1955a, b). Electrical stimulation of the cortex excites the nerve fibers of the superficial layer. The excitation spreads through the branches of the fibers until it reaches the axodendritic synapses where it activates the dendrites (Roitbak, 1950). This interpretation of the origin of the slow negative potential in response to direct cortical stimulation was disputed by Chang (1951a) and Burns and Grafstein (1952), who thought that the slow potential arises without involvement of the fibers in layer I of the cortex, being evoked by an electrical current directly stimulating the dendrite branches. Many facts now indicate that the dendrites cannot be excited electrically, thus making the aforementioned assumption less likely and confirming Roitbak's view. On the other hand, there is some evidence (Ochs, 1956) that elements in the deeper layers of the cortex must be excited if a surface response is to be elicited. Synaptic stimulation of the apical dendrites causes depolarization of the dendrite membrane for 15-20 msec, which reflects the excitatory postsynaptic potential. Isolated stimulation of the dendrites does not excite the corresponding cell bodies (Roitbak, 1953a). The reaction is limited to local excitation of the apical dendrites (Clare and Bishop, 1954; Beritov, 1948b; Beritashvili, 1949). Dendrites a r e functionally different from other parts of the neuron. They are more sensitive to physical injury, to lowering of temperature and anoxia, and to pharmacological agents. Even after a slight injury to the cortex (impaling a capillary 25 p in diameter), the activity of the apical dendrites is blocked for about 20 min (Chang, 1958). These dendrites are not refractory so that the phenomena of temporary summation arise. Summation occurs distinctly in response to high frequency of stimulation. According to Chang (1951b) and Clare and Bishop (1955a), postsynaptic dendrite potentials may be transmitted from the apical branches to the cell body with considerable decrements. The rate of spread through the dendrite is of the order of 0.2-0.6 m/sec. On the other hand, an impulse arising in the cell body apparently does not reach the smaller branches of the apical dendrite. It is therefore reasonable to assume that the individual synaptic neuronal regions (apical and basal) may, in a sense, function separately. For example, while dendrite activity is inhibited, the neuron body retains its capacity to respond to Stimulation (Grundfest and Purpura, 1956). However, there is also evidence pointing to a connection between the processes of depolarization or hyperpolarization of the apical dendrites and facilitation or inhibition of synaptic excitation in the cell body. For

NEURON EXCITABILITY AND DENDRITIC POTENTIALS

17

example, Beritov and Roitbak (cited by Roitbak, 1955) showed that relatively weak cortical stimulation at the rate of 50-103 c/s in the apical dendrite zone evokes a lcng negative potential with simultaneous infiibition of "spontaneous" activity. This is regarded by the authors as inhibition of neuron body activity in the phase of development of persistent depolarization in the dendrite region. Stronger stimulation of the cortical surface may result in a long positive potential accompanied by an increase of "spontaneous" activity. The authors see the mechanism of this interrelationship in polarization of the neurons when their bodies, which ly in the deep layers of the cortex, acquire an electrical charge with a sign opposite to that in the layer of the apical dendrites. In the light of Grundfest's investigations (1957c),it is fair to assume that positive dendrite potentials reflect the hyperpolarization of the dendrite membrane. Under natural conditions, the dendrites may be excited by impulses reaching them via two types of pathways. The impulses reaching the cortex along the classical, specific pathway evoke a negative potential on the cortical surface which reflects postsynaptic depolarization of the dendrites. However, a relatively rapid positive wave is recorded before this surface-negative potential, showing that excitation of the dendrites is preceded by excitation of the elements located in the deep layers (possibly the pyramidal cells in layers II and III). This rapid positive wave is absent with stimulation of the nonspecific pathways because it appears that in this case the apical dendrites are the f i r s t to be excited. Thus, the nature of the cortical potential probably varies with the order in which the axodendritic and axosomatic synapses are excited. The apical and basal dendrites receive different afferent impulses, resulting in a difference in potential arising between these regions of the neuron. The apical-basal gradients may be a factor that modulates the frequency of discharges in the cell body of the neuron. Since the dendrites receive impulses through numerous synapses, it is fair to assume that they perform to some extent an integrating function at the cellular level by appropriately modifying the degree of polarization of the somadendrite membrane. Yet, one cannot rule out the possibility that the branched dendrite network serves to connect the neuron with the environment, to effect the processes of exchange with it. The radial orientation of the apical dendrites facilitates the appearance of synchronous potential oscillations. In the light of Grundfest's data considered above on the electrical nonexcitability of the dendrite membrane, we must assume that synchronization of dendrite activity i s caused by synchronization of the presynaptic processes. There a r e many hypotheses on the mechanism of synchronization of neuronal activity. One of them attributes neuron synchronization to the influence of electrical fields in the environment surrounding the neuron. According to Fessard (1954)) electrical fields in a nervous structure may affect excitability by altering the polarization of the neuron membrane and, as a result, the conditions under which postsynaptic potential arises. This hypothesis is based on experiments in which weak electrical fields created by nervous activity caused synchronization of processes in nervous elements; these fields are the only factor in such synchronization.

18

S T R U C T U R E AND FUNCTION O F NEURONS

For example, impulses with the same rhythm arise in two axons located in an electroconducting environment when only one of them i s stimulated (Arvanitaki, 1942). It has been shown that synchronization of potential oscillations is also possible in cortical structures through purely physical interaction, by electrical fields (Gerard and Libet, 1940). However, synchronization in the cortex is scarcely due to the fields arising around the axons because the latter are partly isolated from the environment by the neuroglia and have contact with the environment only at certain points. Electrical resistance (impedance) of the environment is also of significance in synchronization of activity by the mechanism of electrical fields (Aladjalova, 1960a). Synchronization may likewise a r i s e by simultaneously blockade of excitation of the inhibitory or excitatory synapses. We may posit in this connection the existence of a humoral mechanism of synchronization in the form of active substances appearing in the environment that s e lectively influence various synapses. The cycle of changes in neuron excitability following solitary stimulation i s very brief. However, a state of increased excitability may be maintained by a prolonged inflow of subliminal afferent impulses. For example, constant stimulation of the visual receptor provokes a response both in the visual and in the auditory regions. Changes in the excitability of the cortical neurons may a r i s e as a result of a flow of impulses along the specific as well as the nonspecific ( e . g . , through the diffuse thalamic projections) pathways (Narikashvili, 1958, 1959; Narikashvili and Moniava, 1957), probably through the mechanism of dendrite potentials. A considerable number of phenomena in the cortex a r e closely associated with the dendrite processes. It has been shown, for example, that limited spasmodic discharges may a r i s e after stimulation of the surface layers of the cortex (Morrell, 1960) and spread through the dendrite plexuses (Green and Naquet, 1957). Worth noting are the results of investigations during the past few y e a r s which showed that brief rhythmic stimulation of the nonspecific nuclei of the thalamus (Goldring and O'Leary, 1957) and reticular formation of the brain stem (Arduini et aZ., 1957) evokes a persistent negative potential on the surface of the cortex; this phenomenon is not caused by stimulation of the specific nuclei. An increase in the amplitude of this potential together with higher frequency of stimulation (2.e. the phenomenon of summation, which is especially pronounced in the dendrites) and an assumption that the nonspecific pathways mostly terminate in the dendrites lead all to the conclusion that there is a relationship between a permanent change in potential and the dendrite processes. A similar persistent potential oscillation was obtained in the formation of conditioned reflexes (Rusinov, 1958a; Morrell, 1960). This fact was the basis for the assumption that processes in the apical dendrites play a role in closing temporary connections. These findings were anticipated by Bekhterev who wrote in 1896: "The first cortical layer

NEURON EXCITABILITY AND DENDRITIC POTENTIALS

19

presents in general the most favorable conditions for integrating activity". Recent evidence suggests that brain activity is necessarily integrated by interactions of various kinds and that it cannot be determined solely by the conductivity of excitation along isolated pathways. Examination of even such an elementary interaction as that of two interacting neurons shows how necessary it is to take nonimpulse phenomena into account. For example, axon of cell A reaches the receiving a r e a of cell B and participates in the process of changing i t s excitability; the latter, in turn, varies with the metabolic situation in that area, developing from a local chemical gradient related to the dynamics of metabolism, from the dispersion of active hormonal substances, including those liberated by the afferent endings, and from many other factors. These local chemical processes a r e partly reflected in an electrical phenomenon of an infraslow order. A nerve impulse causes specific changes in a neuron that interact with more diffusely distributed factors in an aggregation of cells. The types of interaction a r e further complicated by the fact that convergent signals reach the neuron chains with their own rhythmic activity. Prolonged changes in excitability a r e determined by the entire complex of these processes. Some of them are electrically manifested in the form of impulses, while others a r e reflected in much slower changes in both electrical potentials and "passive" electrical parameters of tissue. These slow phenomena have received relatively little attention by r e s e a r c h e r s and it is to them that we devote the following pages.

20 CHAPTER II

D.C. POTENTIALS OF THE CEREBRAL CORTEX CLASSIFICATION OF POTENTIALS IN THE CEREBRAL CORTEX BY THEIR DURATION

Biopotentials were once divided into two main types. One included the "resting" potentials (together with the l1injury" potentials); they were measured between the inside and outside of the cell, between injured and intact points on the cell or tissue surface or between different intact points on the surface of the body. The second type consisted of "action" potentials arising during the conduction of a stimulus along a nerve or during cell activity: they could be recorded either from the tissue surface o r intracellularly. "Action" potentials are transient impulses lasting one o r more milliseconds. On the other hand, "resting" potentials are direct-current potentials whose amplitude changes very slowly (in minutes) and these changes are usually arhythmic. It has not yet been conclusively determined which processes or which forms of excitation are reflected in an electroencephalogram. Rusinov's theory (1947, 1954, 1955), derived from the Vvedenskii-Ukhtomskii theory of local excitation, was an attempt to overcome the difficulties. In addition to the traveling wave (action potential) and the electrotonic processes, Rusinov distinguishes a peculiar local excitation which, although a spike potential, spreads through elements of the cerebral cortex. Slow local potentials spread decrementally and last about 10-30 msec. They generally arise in response to presynaptic excitation. The nature and classification of the rhythms of electrical activity were discussed in detail by Beritov (1948a) and Roitbak (1950, 1955), who showed that slow potentials in the spinal cord and brain may be connected with dendrite activity. In recent years the use of intracellular recording of potentials by meansof microelectrodes about 0.1 p in diameter have made it possible to classify cortical potentials with considerable accuracy. The insertion of microelectrodes into elements of the cat cortex enabled Li (1959) to distinguish the following types of phenomena: (1) Insertion of a microelectrode into a cell resulted in a difference in potential reflecting penetration of the electrode through the membrane. A very stable (lasting 40 min) potential of the order of 50-90 mV was subsequently recorded from the cell. Li assumed that the microelectrode here w a s introduced into a nonexcitable cell, probably into a glial cell, because no discharges could be seen on the recording and the potential did not change in response to stimulation of the receptors o r thalamic nuclei or to the action of strychnine. (2) Small "spontaneous" potentials of fluctuating character (Fig. 2, I, B) with an amplitude of 0.8-1.5 mV and duration of 2-7 msec were derived from certain cells. The frequency of these small potentials

CLASSIFICATION O F POTENTIALS BY THEIR DURATION

21

Fig. 2. Classification of cerebral potentials. (I) extracellular recording from neurons of the cat sensorimotor cortex with a general amplitude scale (50 mV) and with different time scales ( L i , 1959); (A) large slow potential regarded as postsynaptic and dendritic; (B) spontaneous subliminal small potentials in the region of the synapse; (C) postsynaptic potential derived from the cell body; (D) action potential of the cell body; ( E ) action potential of the axon; negative deviation upward. (II) electrocorticogram; (A) infraslow rhythmic potential oscillations; (B) with a frequency of 7 osc/min and an aperiodic wave of negative potential; (C) with derivation from the surface of the cat cerebral cortex.

differed from cell to cell; it was of the order of 13 c/s in some, 120 c/s in others. These fluctuations were presumably due to the transmission of subliminal excitatory and inhibitory influences through the synapses, a circumstance according with the hypothesis of "quantum" liberation of the transmission substance (Eccles, 1957). Philipps (1956a) also observed fluctuations of this order when deriving potentials from the pyramidal cell body in the cat cerebral cortex. (3) Slow local postsynaptic potentials (Fig. 2, I, C) with an amplitude of 6-35 mV and duration of 4-7 msec arising in the cell body may be recorded with an extracellular electrode. (4) Rhythmic spike potentials can a r i s e after these postsynaptic potentials, apparently generated by the somatodendritic p a r t of the cell body membrane (Fig. 2, I, D). Duration of such potential: 1.2 msec; frequency: 30-40 c/s; amplitude: 80 mV. (5) Even briefer discharges may be related to the action potentials of

22

D.C. POTENTIALS O F THE CEREBRAL CORTEX

the axons (Fig. 2, I, E). Such brief spike potentials do not show up in an electrocorticogram. According to L i and Jasper (1953), who used extracellular microelectrode recording from the deep layers of the cortex, slow potentials in the electrocorticogram (lasting some 100500 msec) a r e the result of summation of the local processes lasting 15-20 msec associated with the postsynaptic potentials (Fig. 2, 11, A). With extracellular recording, even slower rhythmic potential oscillations can be recorded than the slow local potentials and electrocorticographic potentials called infraslow (Aladjalova, 1956, 1957). Infraslow oscillations derived from the cerebral cortex, as already noted, have an amplitude ranging from 0.2-3 mV and a frequency of two orders: 7-8 and 0.5-2 osc/min (Fig. 2, 11, B). Infraslow oscillations a r e similar to the sinusoidal in form. Infraslow rhythmic potential oscillations a r e to be distinguished from the aperiodic solitary slow shift of potential (Fig. 2, 11, C). A solitary slow shift with an amplitude of several millivolts and duration of about a minute o r more occurs in response to stimulation (Beritov, 1941; Rusinov, 1947; Le%o, 1944a; Burex, 1954a) and is transient. Burex and Koshtoyants (1955) described a case involving the recurrence 5-7 times of such aperiodic shifts of potential after the application of potassium chloride on the surface of the cortex. An aperiodic shift of potential derived from the surface of the c e r ebral cortex may be evoked by frequent rhythmic stimulation of a peripheral nerve (Larionov, 1899; Goldring and O'Leary, 1957), thalamic nuclei (Goldring and O'Leary, 1957), or reticular formation of the brain stem (Arduini et al., 1957) and is regarded as the result of summation of trace processes (Goldring et al., 1958). Thus, the following types of electrical phenomena can be identified in the central nervous system on the basis of recent data: (1) D.c. potentials (including membrane resting potential) with an amplitude of the order of 60-80 mV. (2) Pulse potentials: action potentials lasting 0.2-3 msec. (3) Slow local potentials postsynaptic in origin and lasting 10-20 msec, caused by the processes of postsynaptic transmission. The slow electrocorticographic rhythms evidently develop from these processes. (4) Small periodic potential fluctuations in the synaptic regions. (5) Infraslow potential oscillations (ISPO) with a period of 8-100 sec and an amplitude of 0.5-3 mV. (6) Solitary aperiodic shifts of d.c. potentials lasting 0.5-2 min occurring in response to stimulation. The t e r m "d.c. potential" is a rather artificial combination of the words d.c. and potential. The literature includes attempts to introduce other t e r m s to describe the phenomenon: stationary, stable, persistent, and quasi-steady potentials. We shall use the last t e r m in this book. All these potentials may be directly or indirectly reflected in an electrocorticogram. Prawdicz-Neminski (1913, 1925) was the first to classify the more rapid oscillations in an electrocorticogram. He identified "first-order" oscillations with a frequency of 10-15 c/s, "secondorder" oscillations, 20-32 c/s, and slower oscillations, 0.08-0.5 c/s. H i s work inaugurated the period of oscillographic investigations of

RECORDING O F SLOW POTENTIALS

23

rapid electrical phenomena in the brain. A detailed examination of these oscillations does not come within the scope of our work. RECORDING O F SLOW POTENTIALS

A d.c. amplifier is used to record potential changes, both rapid and slow. The frequency response curve of a d.c. amplifier proceeds horizontally from "zero" frequency and has a steep slope only at fairly high frequencies (2000-5000 c/s). In designing a d.c. amplifier, it is very important to provide for the necessary stability. If direct coupling of stages is used (without bypass capacitors), any change in supply voltage, irregularity in the emission current of tubes, grid-current fluctuations in the first tube, etc. cause changes in voltage at the amplifier output comparable to the changes resulting from the action of the effective signal at the input. One more requirement is important for biological purposes: the absence o r minimum value of the grid currents in the first stage. The reason is that the grid currents may cause polarization of electrodes, which will affect the functional state of the biological structure. The requirements for the absence of zero drift in the amplifier may be satisfied by two different design principles: (1) Modulated amplifiers. A vibrating switch with frequency of 50-100 c/s is used as the input of an a.c. amplifier. Inthis case, the vibratory oscillations are amplitude-modulated by the input signal and amplified. The r e v e r s e operation occurs at the output stage and the potential to be measured is recorded at the amplifier output. Modulation of the c a r r i e r frequency of the amplifier may also be achieved electrically, without mechanical switching. Amplifiers constructed in accordance with this principle may have high gain with complete absence of drift, but the bandwidth is limited. (2) Direct-coupled amplifiers amplify the signal directly and have wide frequency response. To eliminate drift, special circuits are used to obtain high gain (of the order of 100) at the first stage. These amplif i e r s require carefully regulated power supplies. In using microelectrodes a cathode follower should precede the amplifier in order to match the high impedance of the microelectrodes with the lower input impedance of the amplifier. The main requirements of the cathode follower circuit are: high-resistance input, absence of grid currents, operational stability, and amplification close to unity (cathode followers usually have an amplification of less than unity, i.e., they weaken the overall gain of the circuit). Cathode followers have special electrometer tubes with high resistance between grid and cathode and with low interelectrode capacitance. This stage is mounted on an external chassis close to the object (Byzov and Bongardt, 1959). The diagram of the d.c. amplifier that we used for most of the investigations described below is shown in Fig. 3. The stages of the amplifier tire indicated in dashed lines. The rest of the figure contains the c i r cuits of voltage-regulated rectifiers. One.,of them is designed for high voltage (about 900 V) to power the first stage of the amplifier, while

I Cathode current from battery 3-NS-110, 6.3V

BA-6

4.0pF

I

m

Fig. 3. D.c. amplifier (I) with stabilized plate supply (I1 and III); (11) rectifier for 500-1000 V; (III) rectifier f o r 150-300 V. Types of tubes a r e according to Russian standards.

cl

E

RECORDING O F SLOW POTENTIALS

25

the second is designed for low voltage (about 300 V) to power the second stage. The tube filaments a r e heated by storage batteries. A needle millivoltmeter (R = l o o n ) and the 5T loop of an MPO-2 loop oscillograph a r e connected in s e r i e s at the amplifier output. Average gain (with 1 mV input) provides a beam deflection of 12 mm on photographic film. Under these conditions zero drift is virtually absent for 30 min. This circuit is simple and highly stable due to operation of the tubes of the first stage under so-called "starved-circuit" conditions. The choice of a large plate resistor in the f i r s t stage (16 Mn) makes it possible to operate with a very low plate-current and still obtain a gain of 1000 in the first stage. Under these conditions the plate current of the f i r s t tube is several microampGres, the grid-plate transconductance decreases, and internal resistance increases to 40 M n The high input resistance of the tube of the f i r s t stage of the amplifier reduces the grid currents considerably due to the "starved-circuit" conditions and makes it possible to derive potentials in circuits with high internal resistance without a cathode follower. If necessary, however, a cathode follower may be connected at the amplifier input. Voltage a r r i v e s at the screen grid of the first tube from the cathode resistor of the second stage, which serves as a power amplifier. This creates a negative d.c. feedback which helps to regulate the operating conditions of the first tube. The second stage is a bridge-type circuit with closely matched tubes s o that slight shifts in the voltage supply have no affect on output voltage. Gain control i s achieved at the cathode of the power stage. Compensation of the initial potential between the lead-off electrodes is effected by regulating the voltage at the screen grid of one of the tubes of the first stage. The amplifier can boost the potential increment from 100 pV to 30 mV (with appropriate gain control). A shielded chamber is not necessary. The amplifier has a push-pull configuration since most of the measurements a r e made with bipolar derivation. An amplifier designed according to this principle w a s described by Kozhevnikov and Soroko (1956). Attention must be focused on the electrodes when measuring steady and slowly changing potentials. The electrodes should be so chosen that the results of the measurements would not be affected if they became polarized. We tested a variety of electrodes for different parts of the experiments. In acute experiments, we were able to use the classical non-polarizable calomel electrodes that come in contact with tissue through a salt bridge using a wick or through a capillary filled with physiological solution. The resistance of a microcapillary electrode can be reduced by filling it with a three-molar solution of potassium chloride. Wick electrodes a r e inconvenient because the a r e a of contact is diffuse. Their advantage lies in the lack of mechanical stimulation of the brain and in the possibility of maintaining contact while brain tissue is pulsating. Microcapillaries can be used for precise local recordings from different levels of the cortex and individual cells. Depending on the diameter of the microcapillaries (0.1-1.5 p), these electrodes have a d.c. resistance of from 300 k n t o 400 Mn, which may be the source of artefacts (Adrian, 1956). The virtue of this system of

26

D.C. POTENTIALS O F THE CEREBRAL CORTEX

electrodes is the low electrode potential (Kurella, 1958a, b). Chlorinated silver electrodes can likewise be used to record slowly changing potentials in an acute experiment. However, several minutes (2-5 min) must elapse between the time they are applied to the tissue and the s t a r t of recording to permit a determination of the electrode potential. The fact that the conditions of ion equilibrium have developed on the electrode can be judged from the halting of the "drift" of the potential to one of the sides. Recording by chlorinated silver electrodes is also possible from skull skin. In this case, a strip of filter paper or wool moistened with physiological solution should be inserted between the silver plate and the skin. If potentials are to be derived from an exposed cortex, one should bear in mind the fact that the electrodes must not be allowed to exert more than 1 - 2 g/cm2 of pressure on the brain. This amount is essential for a good contact but does not cause ischemia in the subjacent tissues. Electrode pressure on the cortex may alter the functional state of nervous tissue, a fact that must be remembered when implanting plate electrodes. In chronic experiments, the material for the implanted electrodes may be highly varied. We have successfully used silver, platinum, nichrome, and tungsten. Very rigid material, e g . , nichrome, should be chosen for subcortical electrodes 10-20 in diameter. Slowly changing potentials can be recorded either bipolarly or unipolarly. The electrodes can be set to within several tens of microns of each other in the case of bipolar derivation. This method permits the achievement of maximum local recording. With unipolar recording, the indifferent electrode and its location must satisfy several conditions, principally the invariability of potential near it. Many of the methods described in the literature do not meet this requirement. In experiments with improperly placed electrodes, the curve reflecting the brain currents will have superimposed skin potentials, muscular currents, contact noises, etc. An ideal indifferent electrode is one i r r a e r s e d in a basin of water several tens of centimetres away from the body of the animal submerged there. The ear lobe, for example, is a good place for the indifferent electrode in recording an EEG. However, d.c. potentials require more precise conditions for measurement in order to avoid e r r o r s in interpreting the recording. The indifferent electrode is best implanted chronically because if applied to freshly dissected tissue, it results in fluctuation of the electrode potential due to ion shifts (injury potentials, etc.). The indifferent electrode was placed either on a coagulated portion of nervous tissue or in physiological solution that washed the object being measured. Examination of the physicochemical phenomena on the electrodes (Frumkin et al., 1952) shows that they consist of more or less long standing double ionic layers whose discharges exchange with tissue discharges. This process is steady and it cannot be rhythmic. However, the pre-electrode ionic layer will change with changes in temperature and medium acidity, thus affecting the drift of potential in one direction or another. When measuring slow potentials, one must always be aware of the presence and nature of the pre-electrode processes.

DIFFERENCE IN POTENTIALS

27

DIFFERENCE IN POTENTIALS BETWEEN DIFFERENT POINTS AND AT VARIOUS DEPTHS O F THE CEREBRAL CORTEX

Dubois-Reymond w a s the f i r s t to discover the electrical properties of the brain, but for a long time no one continued his observations. More than 30 y e a r s later electrical phenomena in the brain were described independently by Danilevskii (1891) and Caton (1891), who initiated the use of galvanometric investigation of currents in the cerebral cortex of animals. Sechenov's work (1882) was a prerequisite of further elaboration of the problem of cerebral electrogenesis. Sechenov showed that excitation of the frog spinal cord in response to stimulation of nerves is associated with aperiodic negative current oscillation. H i s discovery of "spontaneous" periodic oscillations derived from the medulla oblongata w a s particularly important. A similar pattern of electrical phenomena in the brain and spinal cord was also noted by Vvedenskii (1884) and Verigo (1889). Some investigations appeared in the 1890's (Beck, 1890) that demonstrated the existence of negative current oscillations in an excited region, which were prevented, however, by the u s e of anesthetics. Workers in Bekhterev's laboratory used a galvanometer to determine the localization of the auditory (Larionov, 1899) and visual (Trivus, 1900) a r e a s in the cerebral cortex in response to stimulation of the receptors. Another worker in Bekhterev's laboratory, Kaufman (1912), made a detailed investigation of the topography of the potential sign on the surface of the hemisphere. Kaufman provided experimental proof of the possibility of recording currents from the skull and showed that there is a difference in potential between the frontal and occipital r e gions of the hemispheres, which increases in asphyxia and decreases when blood pressure is lowered. External influences on the cortex (cauterization near the electrode) made the underlying area electronegative. A difference in electrical potentials between various regions of the hemispheres w a s observed several minutes after an animal died. Besides aperiodic shifts in potential, Kaufman also found "spontaneous" oscillations of currents. According to Aleksanyan and Demirchoglyan (1955), flashing a light just once into the eye caused an aperiodic potential wave (sign not mentioned) and indistinct changes in the slow "spontaneous" oscillations (recorded with a galvanometer) in the cerebral cortex. These reports on the existence of very slow, essentially aperiodic electrical processes in the brain were not immediately followed up because the galvanometric methods were supplanted by electronic techniques, which were the first to utilize a.c. amplifiers. Recently, however, improved methods of amplifying direct current stimulated the appearance of a number of investigations on d.c. potentials in the brain that can be carried out with accurate quantitative evaluation of the results. The observations deal mainly with aperiodic shifts of potential in one direction. The results of the measurements in these experiments were determined by the sum of the potentials created by many sources. The electrical field that arises around a dipole (Fig. 4) varies with

28

D.C. POTENTIALS OF THE CEREBRAL CORTEX

Fig. 4. Electrical field around a dipole. Solid curves: lines of force; dotted curves: equipotential lines.

the discharge of the poles of the dipoles slightly separated, the dielectrical constant, and electroconductivity of the medium. In the case of a conducting medium, the lines of the current pass along the lines of force. In a medium with heterogeneous impedance, the lines of the field converge in the less conducting layer and readily bend round the conducting layers; the current lines a r e distributed in reverse fashion. An electrical field was also found around a nerve (Burr and Mauro, 1949). It extended for 6 mm from a myelinated nerve. Lorente de N6 (1939) studied changes in the electrical field formed in Ringer's solution around the frog sciatic nerve while waves of excitation were coursing in it. The potential near the unexcited p a r t s of the nerve was positive relative to the distant indifferent electrode but negative around the excited parts. Kostyuk (1959) thinks that electrodes placed in a volume conductor near an excited cell (extracellular recording) lead off a difference in potential in the surrounding electrolyte that is created by ring currents from the excited p a r t of the cell surface to the unexcited part. Some components of the difference in potentials in the brain may be due to the presence of neuron segments with varying polarity, while others are due to the difference in ion concentration at the anatomical boundaries (e.g., at the blood-brain barrier), etc. According to the most widespread viewpoint (Libet and Gerard, 1941; Goldring and O'Leary, 1951b), the source of electromotive force with stable differences in potentials i s the actual population of the neurons due to their polarization and identical (radial) orientation. This is proved by the relationship existing between the magnitude of the difference in potentials and neuron function ( e . g . ,potential may shift in the cerebral cortex as a result of summation of trace potentials arising during neuron activity). According to another viewpoint (Tschirgi and Taylor, 1958), it is possible for a stable difference in potentials to develop due to distribution of ions on the blood-brain barrier. The author's hypothesis is supported by the presence of a steady difference in potential between an electrode inserted in a vein and the surface of the cerebral cortex, the latter being 1-5 mV more negative

DIFFERENCE IN POTENTIALS

29

relative to the blood in the jugular vein. This difference in potential, like that between two points on the cortical surface or between a point on the cortical surface and a ventricle, changes under deep anesthesia, in response to pain, etc. A several millivolt difference in potential arises between different points of the cerebral cortex, the anterior regions being more negative relative to the posterior and the central zone being more negative relative to the lateral portions of the hemisphere (Gerard and Libet, 1940). Points in homologous regions of the cerebral cortex are relatively isopotential (Kernpinsky, 1954) or their potential differs only by 1-2 mV. Different points on the cortex have a different potential relative to the subjacent portions of the hemispheres. The difference in potential between the pia mater and the lateral ventricles of the brain is of the order of 1-7 mV. According t o some authors (Goldring and O'Leary, 1951a), the cortical surface is positive relative to an indifferent electrode in most cases; according to other authors, it is now positive, now negative (Kempinsky, 1954). A difference in potential between the pia mater and the lateral ventricle occurs in both man and monkey (Goldring et al., 1950) and it can be measured by electrical stimulation or by injecting metrazol or insulin. Transcortical measurement of potentials has become very popular in recent y e a r s (Gerard and Libet, 1940; Goldring and O'Leary, 1951a, b, 1957). A shift of potential in these experiments is interpreted as a change in polarization of the pyramidal neurons. This accounts f o r the relationship observed between the nature of spontaneous or induced activity, on one hand, and the shift in quasi-steady potential, on the other. Measurement of the quasi-steady potential at different levels of the cortex by means of non-polarizable extracellular microelectrodes (Aladjalova and Koshtoyants, 1957) shows that the deeper the microelectrode is implanted in the cortex, the more negative the potential becomes, the zone of maximal negativity being approximately at the level of layer V-VI (Fig. 5). Our experiments were performed on the sensorimotor region of the rabbit cortex with unipolar recording; a small plate resting on a coagulated a r e a of the other hemisphere served as the indifferent electrode. The initial difference in the poten-

Fig. 5. Distribution of potential at different levels of the rabbit cerebral cortex (3 experiments). Y-axis: difference in potentials between the surface of the cerebral cortex and the deep layer of the cerebral cortex in which the microelectrode was implanted; X-axis: depth to which the microelectrode was implanted. Thickness of the cerebral cortex 2 mm; the microelectrode w a s later implanted in the white substance.

30

D.C. POTENTIALS OF THE CEREBRAL CORTEX

tial between the recording microelectrode (10 p in diameter) and the indifferent area on the adjacent hemisphere was compensated in the recording apparatus and taken for zero. Thus, the potential in different layers of the cortex w a s measured in relation to the potential of the coagulated area of the surface of the symmetrical hemisphere. This s e r i e s of experiments was performed on 1 2 rabbits and repeated several times in each animal. The zone of maximal negativity w a s found in all the experiments in different cortical regions in the lower part of the diameter of the cortex with a spread of + 0.26 mm. When the microelectrode was inserted again in the same region of the cerebral cortex, the zone of negativity was recorded with a spread of + 0.1-0.15 mm. The white substance near the gray matter w a s more negative than the surface of the cortex. When the microelectrode was inserted more deeply into the white substance, the potential recorded w a s more positive. Thus, there is a level of maximal negativity at a depth of 1.4-1.6 mm, which touches layer V of the cortex or is found somewhat below it. An electrical field of the opposite sign arises, therefore, in the upper and lower layers of the cortex (Fig. 6). It was Adrian who first advanced (1936) the hypothesis of a dipole existing in the cortex oriented radially and with a positive pole in the region of the apical dendrites. Distribution of potential around a group of active neurons (in a conductihg medium) will necessarily vary according to the relation of their active and resting parts, i . e . , whether they are depolarized or repolarized (Burns, 1958). The zone of maximal negativity is found in those layers of the cortex which, according to Sholl ( c j . p. 6), contain the highest density of cells. Yet it is right here, in layers V-VI, that the veurons degenerate completely 6-8 weeks after chronic transection of the connections with the subcortical structures whereas the neurons of the upper layers are preserved (Burns et al., 1957). Another important feature of neurons in layers V-VI is perhaps their noninvolvement in spike activity following excitation of an isolated slab of cortex, whereas the neurons in layers TI, III, and IV, classified by Burns (1958) as type B neurons, exhibit spike activity for 0.5-7 see after single stimulation. Analysis of these facts suggests that while some neurons exhibit spike activity, others create prolonged potentials. Several y e a r s ago a subdivision of cells into these same categories w a s experimentally demonstrated for the retina of the fish eye (Grundfest, 1958). The upper and lower layers of the cortex differ in many respects: intensity of metabolic processes (Dixon, 1949b, 1953), electrical activity (Sarkisov and Livanov, 1933), nature of development in ontogenesis, etc. Rabinovich (1958) found that the activity of the upper and lower layers does not develop equally under the influence of external stimuli. Knipet (1955) showed that the rhythm of receptor stimulation is reproduced first in the electrical response of the upper layers, then in the response of the lower layers. Modern findings imply that the electrical phenomena which arise in layers V and VI may be associated with the function of the nonspecific formations of the brain. Thus, the evoked potential (in response to stimulation of the specific afferent nuclei) has a different electrical sign in

DIFFERENCE IN POTENTIALS A

0

A A

I 2

3 4

Fig. 6. Comparison of distribution of the quasi-steady potential with structural characteristics of different layers of the cerebral cortex. (A) schematic representation of the cerebral cortex after Lorente de N6 (I-VI) layers of the cortex; 1-2 = specific afferent fibers from the thalamus, 3-4 = nonspecific afferent fib e r s , 5 = axodendritic endings of nonspecific afferent fibers, 6 = axosomatic endings of specific afferent fibers, 7-8 = association fibers, 9 = large pyramidal cells with long axon; (B) cytoarchitectonics of the motor zone of the rabbit cerebral cortex (only the nerve cell bodies are stained); (C) distribution of the quasi-steady potential at different layers of the cortex; (D) direction of elect r i c a l field lines of force in upper and lower layers of the cerebral cortex.

the upper and lower layers of the cortex (Roitbak, 1955; Li et al., 1956a). It has been noted (in experiments on cats) that this "inversion" of the sign occurs at a depth of 0.6-0.9 mm, z.e., at the level of layers III and IV. The sign of the recruiting potentials (Li et al., 195613) may change at the level of layers V-VI. These potentials appear in the cortex in response to stimulation of the intralaminar "nonspecific" nuclei of the thalamus. The nonspecific afferent fibers (Fig. 6) described by Lorente de N6 give off collaterals in all the layers of the cortex. Their endings are localized (Chang, 1952b) on the dendrites. A number of authors (Powell and Cowan, 1956), however, were unable to find anatomically "nonspecific" pathways. The fact that the most negative persistent potential is recorded at the level of layers V-VI, is ground for believing, when considered in connection with the data set forth above, that this potential is associated with the neurons that receive transmission along the nonspecific pathways. The existence of a zone of negative potential at a certain depth of the cortex may create a steady electrical field of different direction in its upper and lower parts. For example, the vector of the electrical field in the upper layers is directed from the apical dendrites to the

32

D.C. POTENTIALS OF THE CEREBRAL CORTEX

cells in layers IV and V. In the lower layers, on the other hand, it is directed from the white substance to layer V. Extracellular currents may arise in accordance with this wave in a conductive medium, SIGNIFICANCE O F EXTRACELLULAR CURRENTS IN NEURON FUNCTION

Elucidation of the function of extracellular currents started with experiments involving the application of artificial polarization from an external source to different divisions of the brain. Scheminzky et a l . (1936) demonstrated that descending currents in amphibians and fish (the head is more positive) resulted in a delayed motor reaction, while ascending currents caused an increase in activity (Vassilyev (1939), Galkin (1953), and others). Lapitskii (1948) succeeded in changing the threshold of convulsive activity in warm-blooded animals by applying a positive potential to the brain. Smirnov (1950) used polarization of the brain to change conditioned activity. Goldring and O'Leary (1951a) induced polarization of the rabbit c e r ebral cortex relaWe to another electrode inserted into the hypothalamus. Negative polarization caused a drop in activity of the cortex, whereas positive polarization (3 mA, 30 sec) produced paroxysmal discharges. This contradicted the findings of the earlier authors, which could be largely ascribed to the different methods of applying the polarization current. The role of artificial polarization was further investigated by Rusinov (1953, 1958a, b) and his coworkers (Novikova et al., 1952; Novikova and Farber, 1956; Sokolova and Khon Sek Bu, 1957; Pavlygina, 1959) and by Kostyuk's detailed analysis (1959) of the spreading of artificial currents in multicellular structures. Rusinov and co-workers found that positive polarization of the sensorimotor a r e a of the cerebral cortex facilitates execution of the conditioned defense reflex to sound. The existence of a persistent drop in potential may have a prolonged (from 40 min to 2-3 h) effect of electrotonic character on interneuronal activity. A direct current in the cortex gives rise to a fixed focus of excitation which acquires dominant properties and creates optimum conditions for closure. When the power of a polarizing current is increased, the dominant focus turns into a parabiotic one and the closure is extinguished. A dominant focus may also be formed in the hypothalamus where it persists for several days after the current is switched off. These investigations show that potentials of quasi-steady origin are significant for higher nervous activity in animals. Artificial polarization, which creates a negative charge on the surface of the cerebral cortex and a positive charge on an electrode in the ventricle (frog brain) produces much longer trace effects (sometimes about 1 h) than does polarization in the reverse direction (positive on the surface). In the latter case, the difference in potentials returned to normal within 10 min (Libet and Gerard, 1941). Polarization in a caudorostral direction likewise changes the potential between the brain surface and ventricle. Strong currents a r e l e s s effective than weak ones. According to Gerard and Libet (1940) and Libet and

INFLUENCES ON THE QUASI-STEADY POTENTIAL

33

Gerard (1941), ascending currents (positive on the surface relative to the ventricle) have an activity-exciting effect while descending currents (negative on the surface) inhibit activity. This is essentially the conclusion that Burns (1954) reached after investigating isolated slabs of cortex. Artificial polarization of a small portion of the surface of the cortex w a s induced by a microelectrode. A descending current of about 100 pA (positive polarization of the surface) excited the cortical neurons; this phenomenon did not appear when the direction of the current was reversed. On the other hand, when the microelectrode was implanted to a depth of 1.2 mm, i.e., into layers V-VI, excitation arose only with negative potential on the electrode, failing to arise with positive potential, Rhythmic stimulation of isolated slabs of cortex caused oscillating excitation of the neuron populations. However, application of negative polarization to the cortical surface at this moment or application of positive potential as f a r down as layers V-VI immediately halted the activity of the s t r i p induced by stimulation. Recording of potentials from individual cells by means of microelectrodes showed (Burns et a l . , 19572 that different neurons may r e act differently to polarization. For example, there are cells whose discharges cease upon polarization by an ascending current. On the other hand, there are cells which begin to show discharges only when weak polarization is applied by an ascending (positive on the surface} current. These experimental data provide convincing evidence that weak extracellular currents may influence the activity of the individual neurons, although in different ways for neurons of different types. It is hard to believe that the mechanism of influence of extracellular currents i s related to their effect on the axons because axons in the cerebral cortex are oriented in all possible directions and their selectivity for an ascending o r descending current is incomprehensible. The effect of polarization on the dipole "apical dendrites-cell body" is more likely. The region of the apical dendrites, as we pointed out above, is more positive than the cell body. This dipole creates a descending curr e n t in the upper layers. Artificial positive polarization increases this current, which, in turn, increases the excitability of the cell body. We must emphasize that positive polarization of the cortical surface increases excitability and this is specific for the cell structure of the cerebral cortex and cannot take place in such an elementary excitable system as the nerve fiber. Positive polarization of the latter blocks conduction. This too does not take place on the cell body membrane because an ascending current causes hyperpolarization of the membrane. These facts a r e grounds for believing that "spontaneous" changes in the potential gradient in the cortical structure may influence electrical activity of the neurons. Consequently, infraslow potential oscillation may well be such a factor too.

,EFFECT O F EXTERNAL INFLUENCES ON THE QUASI-STEADY POTENTIAL

The difference in potential between two points on the cortex changes after stimulation of the cerebral cortex, receptors, or subcortical structures, rhodification of brain metabolism, changes in blood circu-

34

D.C. POTENTIALS O F THE CEREBRAL CORTEX

lation, etc. Impairment of the oxygen supply is a particularly potent factor. In 1912, Kaufman noted that compression of the trachea changes the potential of the cerebral cortex by 1-2 mV, at f i r s t in a positive direction (temporary phase), then in a negative direction (a phase lasting several minutes at the beginning of which one can also observe depression of the more rapid electrical activity of the brain). When the carotid artery is compressed, the shift in difference of the potentials between the pia mater and the white substance i s less than that between the white substance and the ventricle (Kernpinsky, 1954). On the other hand, dendrites and neuron bodies a r e known to be more sensitive than axons to oxygen deficiency. The marked shifts of potential in the white substance during anoxia are presumably due not s o much to the nature of the nerve elements as to the peculiarities of the blood supply or to the metabolism of the neuroglia. Determination of potential at different depths of the cerebral cortex (Le%o,1951) shows that during anoxemia the superficial layers of the cortex first become negative and later the deep layers. Different regions of the cortex (motor and visual) likewise differ in sensitivity to anoxemia. The potential of the cortical surface in warm-blooded animals is positive relative to the ventricle (Kempinsky, 1954; Goldring and O'Leary, 1951a) and to the white substance (Aladjalova and Koshtoyants, 1957), but negative relative to the potential of the blood in the external jugular vein (Tschirgi and Taylor, 1958). The difference in potential between nervous tissue and blood decreases in barbiturate narcosis, Inhalation of nitrogen causes a shift of potential of about 15 mV in a negative direction. On the other hand, inhalation of CO, changes it in a positive direction initially; during complete asphyxia the cortical potential relative to the blood becomes 10 mV more negative. This shift p e r s i s t s for about 30 min after cardiac standstill and slowly decreases to z e r o in 1-2 h. Persistence of the potential at a time when the neurons have ceased to function seems to suggest the lack of any connection between the quasi-steady potential and the activity of the nervous system. However, Bures (1954a) showed that quasi-steady potential after the appearance of signs of death r e mains only as long as resuscitation is possible. This testifies to the presence of metabolic processes which p e r s i s t for a,long time after cessation of the inflow of oxygen and which make it possible for functions to be restored. Consequently, the quasi-steady potential is closely connected with vital metabolic processes. Bures' ontogenetic investigations (1957) showed that the steady difference in potentials of the cerebral cortex appears in young rats only after they are 20 days old. This seems to be further evidence of the relationship between potential and structure. A change in the concentration of hydrogen ions in the blood (injection of a solution of hydrochloric acid) and in the fluid surrounding the neurons (interstitial fluid) influences the blood-brain difference in potentials (Tschirgi and Taylor, 1958). However, an injection of potassium chloride into a vein or spinal canal changes the potentials in the opposite direction. These facts suggest the source of the quasi-steady po-

INFLUENCES ON THE QUASI-STEADY POTENTIAL

35

tential to be at the "blood-brain" barrier. Such a difference in potential may arise, for example, at the boundary between the blood vessels and glia. Tschirgi,and Taylor (1958) believe that the glia i s responsible for the transfer of inorganic electrolytes between plasma and interstitial fluid and that it possesses, in addition, a secretory function; the quasisteady potential may partly reflect these secretory processes. All the observed changes in potential obviously cannot be reduced to a single source of electromotive force. At least three mechanisms of their origin can be distinguished: activity of nerve elements, local metabolic shifts, and ion gradient at different anatomical b a r r i e r s , the blood-brain b a r r i e r in particular. A change in potential on the cortical surface spreads to the adjacent areas. Waves of steady potential have been most thoroughly studied in connection with local stimulation of the cortex (Lego, 1944a, b). A nascent aperiodic wave (electronegative) with an amplitude of several millivolts and duration of 0.7-2 min spreads to adjacent points on the cortex at the r a t e of about 3 mm/min. The appearance of an electronegative potential wave is accompanied at first by a brief burst of activity, but within half a minute progressive depression of both rapid and slow electrical activity s t a r t s (Aladjalova, 1956b). In an area encompassed by depression (a 5-8 mm s t r i p of cortex), primary responses to stimulation of specific receptors a r e inhibited, strychnine discharges do not occur, and conditioned reflexes disappear for 30-60 min (Burex and Burezova, 1958), i . e . , several indicators show that the cortex in this a r e a is in a state of depression, which, however, is reversible. The depression lasts only 20-30 min after which the negative wave disappears. Electrical activity resumes within 10-15 min while the infraslow potential oscillations a r e restored in 20-30 min. The phenomenon of an aperiodic wave of negative potential may also be produced in an isolated slab of cerebral cortex (Burns, 1954). The mechanism of spread of an aperiodic wave is still obscure. It apparently spreads extrasynaptically because it is very slow; moreover, hypoxia and anesthesia do not interfere with the spread (Sloan and Jasper, 1950). An electronegative wave is transmitted through the line of an incision in the presence of simple physical contact between portions of the c e r ebral cortex. Activation of the neurons is the first stage in the origin of an aperiodic wave. One can observe in this stage a shrinkage of the cortex, decrease in oxygen p r e s s u r e therein, and constriction of the pial blood vessels (Van Harreveld and Stamm, 1953). These facts suggest that depression may result from vascular chahges, while the mechanism of its spread is related to the release of chemical transmitters acting on the nerve plexuses around the blood vessels (Roitbak, 1955). However, MacIntosh and Oborin (1953), using the method of vividiffusion, were unable to detect any increase in the amount of substances released during depression. The sign of the potential on the surface of the cerebral cortex plays an important role in the transmission of depression. For example,=tificial polarization (Grafstein, 1956b) of the surface during the formation of an aperiodic electronegative wave halts the spread of depression. The following pattern has been observed with tangential placement of

36

D.C. POTENTIALS O F THE CEREBRAL CORTEX

both polarizing electrodes. If a charge that is negative relative to an electrode located close to the parietal region is supplied to an electrode placed close to the frontal region, the r a t e of spread of an aperiodic wave and its m p l i t u d e decrease, The decrease occurs under both polarizing electrodes. If the phenomenon of depression w a s associated with vertically oriented neurons (e.g., pyramidal neurons), the effects under the polarizing electrodes of different sign would be different because the polarity of the dipole under the anode and cathode would change in the opposite direction. It is conjectured that tangentially oriented polar elements, whose existence is still unknown, a r e more likely to play a role in this phenomenon. The author is thus compelled to assume that cations (potassium) are released and move along the surface of the cerebral cortex. There are no clear-cut experiments to confirm this assumption. Chang (1958) thinks that an aperiodic potential wave is an injury potential which can be a transmissive factor and cause depression in the adjacent regions where injury potential then arises, etc. This mechanism can account for the transmission of depression through the line of an anatomical incision. CHANGES IN THE QUASI-STEADY POTENTIAL IN RELATION TO THE ACTIVITY IN DENDRITES

Beritov and Roitbak (cited by Roitbak, 1955) showed that relatively weak stimulation of the surface of the cortex at a frequency of 50-100 per sec evokes prolonged negative potential of nonoscillating character, the effect of summation of dendrite potentials. The amplitude of this potential (an aperiodic wave in our classification) increases with the frequency of stimulation. This aperiodic wave reaches a maximum 1-1.5 sec from the start of tetanization of the cortical surface and weakens after 3-4 sec. Spontaneous activity is inhibited at this time, i . e . , excitation of the dendrites, according to the authors, inhibits functioning of the neurons. The inner layers of the cortex are meanwhile polarized positively. The authors believe that activation of a dendrite plexus in layers I and 111 causes positive polarization in the layer of the pyramidal neuron bodies, thus inhibiting spontaneous activity. Such is the mechanism of inhibition according to Beritashvili's dendrite hypothesis (1953). The quasi-steady potential of the cortex changes whether excitation travels along the specific or nonspecific pathways. Thus, Goldring and O'Leary (1951a, b, 1954), using transcortical derivation (cortex-ventricle), recorded a change in potential of 0.8-1 mV in a positive direction after a primary response of the rabbit cerebral cortex to stimulation of the visual receptor. A more prolonged change in a positive direction was observed during paroxysmal activity caused by repeated stimulation of the lateral geniculate body. The development of spontaneous paroxys mal activity likewise corresponded to the change in potential in a positive direction. The application of strychnine (0.1%) resulted in increasing the aperiodic wave in response to stimulation, whereas the application of cold, which affected only the outer layers of the cortex,

CHANGES IN QUASI-STEADY POTENTIAL

37

decreased the aperiodic wave, thus suggesting a connection between the potential and the processes in the superficial elements of the cortex, Stimulation of the intralaminar thalamic nuclei is known to create recruiting potentials in the cortex. It has been found that trace effects a r e summed after each response so that a marked aperiodic wave a r i s e s (Goldring and 0'Leary, 1957). When the frequency of stimulation is increased, summation for this interval of time has a considerable effect. Stimulation of the brain stem reticular formation also causes a change of several millivolts in the quasi-steady potential of the cortex in a negative direction (Arduini et a l . , 1957); the greater the frequency of stimulation, the larger the amplitude of the aperiodic wave. The authors believe that a gradual component of the recruiting potential is created by the dendrites and that the time between the succeeding stimulating pulses i s insufficient to r e s t o r e the dendrite membrane. Therefore, summation of the trace potential s e e m s to occur. Like the effect of stimulation of the intralaminar thalamic nuclei, the effect of stimulating the brain stem reticular formation is thought to be caused by nerve pulses reaching the dendrites through the synapses. However, Ingvar's experiments (1955b, c) on isolated slabs of cortex showed that stimulation of the reticular formation changes the activity in the strip, i . e . , it acts humorally. Therefore, Goldring and O'Leary's theory of the mechanism of the reticular formation influence on changes in the quasi-steady potential in the cortex is not the only possible one. The quasi-steady potential has recently attracted the attention of investigators due to the fact that it changes during the process of conditioned reflex formation. Rusinov (1958) held that the mechanism of conditioned reflex formation is associated with a change in potential which initially has the nature of a generalized phenomenon, but then becomes concentrated in a zone specific for a particular stimulus. An aperiodic negative wave of quasi-steady potential can be evoked by a stimulus that has become conditioned, although this stimulus p r e viously caused no change in potential. In Morrell's experiments (1960), an aperiodic negative potential wave in the central reg{on of a rabbit cortex was evoked by stimulation of the intralaminar thalamic nuclei. After 30 combinations of a sound signal and stimulation of the thalamus, isolated presentation of the sound evoked the same negative quasi-steady potential wave. The possibility of conditioned reflex formation of this wave, in Morrell's view, confirms the hypothesis that closure of conditioned reflexes at the cortical level takes place in the upper layers of the cortex. This hypothesis is based on that of Goldring and O'Leary (1957) who posited a relationship between an aperiodic quasi-steady potential wave and summation of dendrites responses. The fact that the quasi-steady potential changes in the course of conditioned reflex activity has heightened the interest of investigators in d.c. potentials in the brain.

38

D.C. POTENTIALS O F THE CEREBRAL CORTEX CONCLUSIONS

Brain potentials may be subdivided into spike, slow, infraslow, aperiodic, and steady. Spike potentials reflect spreading excitation of the axon o r cell body and they may be rhythmic. Slow potentials reflect local depolarization o r hyperpolarization (in response to transsynaptic influence) on that part of the neuron membrane which is electrically unexcitable, including the dendrite membrane. Infraslow rhythmic potentials reflect periodic changes in potential in a local portion of the cortex, possibly on the membrane of certain nerve elements, due to metabolic changes. Aperiodic waves reflect a one-time change in potential in some region of the cortex in response to a specific influence and they may be the result of summation of slow trace processes. Steady potentials a r e recorded from different parts of the brain; their sources may be ion gradients created by the diversity of metabolic processes in different structures, different polarity of components of the neuron, and difference in ion concentrations at anatomical boundaries, notably the blood-brain b a r r i e r . Steady potentials a r e the result of metabolic gradients, yet they may also be the cause of changes in the latter ( e . g . ,with spreading depression of electrical activity through the hemispheres). The deeper the microelectrode is implanted in the cerebral cortex, the more negative the potential is relative to the potential of the pia mater. The potential is maximally negative in layers V-VI. This level coincides with the region of greatest concentration of the cells that a r e connected with the nonspecific structures of the subcortical structures. Thus, the upper and lower layers of the cortex have an electrical field of the r e v e r s e direction: the developing extracellular currents flow from the layer of the apical dendrites to the layer of the neuron bodies. Artificial passage of a weak current through the cortex influences its function: descending currents excite the cell bodies of the deep-lying neurons while ascending currents inhibit them. With a descending current, the potential of the cortical surface i s more positive than the potential of a deeper layer, in which excitation may also originate. In the case of natural excitation of the apical dendrites, an electrical field of the same configuration is created as in the case of artificial electronegative polarization of the cortical surface, during which the deep layers become inhibited. However, this rule does not apply to all the nerve elements of the cortex, for some a r e excited (i.e.,they produce discharges) by currents of the r e v e r s e direction. Extracellular currents influence neuron excitability apparently by hyperpolarizing or depolarizing the neuron membrane. The quasi-steady potential of the cortex is also noteworthy because it can be changed by artificial stimulation of different nerve structures and by conditioned-reflex activity. An aperiodic potential wave arising in response to stimulation of certain structures can be elicited by conditioned stimulation. It is regarded as the result of summation of dendrite responses.

39 CHAPTER I11

INFRASLOW RHYTHMIC POTENTIAL OSCILLATIONS IN THE CEREBRALCORTEX

By "infraslow" (Aladjalova, 1956b, 1957) rhythmic oscillations we mean the rhythmic potential oscillations that a r e sinusoidal in appearance (Fig. 7). with a frequency of 7-8 osc/min and an amplitude of 0.3-0.8 mV (A*-rhythm] and with a frequency of 0.5-2 osc/min and an amplitude of 0.5-1.5 mV (B*-rhythm]. Infraslow oscillations differ from the aperiodic min

Fig. 7 . Infraslow rhythmic changes in a recording from the motor cortex (electrodes 1-2, 5-6). (A) = A-rhythm; (B) = B-rhythm; (C) = combination of A- and B-rhythms. Recording made in different experiments without anesthesia. Bottom left: diagram of position of electrodes on the rabbit brain (cf.text).

potential oscillations described in the preceding chapter in that they a r e rhythmic oscillations that may follow one after the other for several hours or days. An infraslow rhythm a r i s e s in another electrochemical system than the other changes in the potential of the brain. Although infraslow potentials arise independently of the more rapid electrical activity, there seems to be a connection between them (cf.Chapter IV). Infraslow rhythmic potential oscillations are also characteristic of the human brain.

INFRASLOW POTENTIAL OSCILLATIONS IN THE HEMISPHERES OF A WAKING ANIMAL

A recording method using macro- and microelectrodes

Infraslow potential oscillations (ISPO) can be recorded both in chronic experiments with implanted electrodes and in acute experiments. In most chronic experiments, the oscillations were recorded from the surface of the rabbit cerebral cortex by means of silver or platinum electrodes in the form of round discs 2-4 mm in diameter. Pairs of

* Due to the difficulty of translating the Russian t e r m s "sekundnyi, minutnyi and chasovoi ritml', the translator preferred to use the terms "A-, B- and C-rhythm".

40

INFRASLOW RHYTHMIC P O T E N T I A L OSCILLATIONS

electrodes were placed on the frontal and occipital regions of the cerebral cortex. The centers of the discs were 5.5-6 mm apart. Electrodes a r e implanted through trepanations in the skull under local or general anesthesia under sterile conditions with subsequent use of penicillin. The electrodes must be implanted at least 10 days before an experiment. They a r e secured to the skull with a sleeve inserted in the trepanation and dental cement. During the experiment the animals with implanted electrodes need not be anesthetized or immobilized in any special way. The only requirement i s a properly fitting head holder so that the animal is not irritated. The choice of material for the electrodes is particularly important in acute experiments, where nonpolarizable electrodes are always preferable, though not essential. Two types of nonpolarizable electrodes can be used. Capillary nonpolarizable electrodes (microelectrodes) a r e used in acute experiments with an exposed cortex. The microcapillary is filled with physiological solution and connected to a nonpolarizable Dubois-Reymond system. In recording from the skin overlying the head, use is made of chlorinated silver electrodes with a sheet of filter paper saturated with physiological solution. A cotton wick moistened with physiological solution should not be used because the amount of surface in contact with the object will vary with the a r e a over which the solution spreads and it may change if ther? is additional moistening, thus altering the recording conditions. Recording of an ISPO may be bipolar or unipolar. The distance between the electrodes in bipolar recording may be quite short (about 0.1-0.2 mm in microelectrode recording). If differences in potential a r e to be recorded from a unipolar lead, particular attention must be paid to the position of the indifferent electrode due to the possibility of artefacts appearing (skin potentials, contact phenomena, proximity of large pulsating blood vessels, etc.). The indifferent electrode should be nonpolarizable. Depending on the aim of the investigation, unipolar recording may present certain advantages because a comparison of the potential in a given p a r t of the brain with a point whose potential is virtually unchanged makes it possible to relate any shifts observed to changes in potential in that particular portion of the hemisphere. When recording an ISPO, one should not use grid currents in the input stage of the d.c. amplifier. Grid currents as they pass through the object being investigated create additional artificial polarization, which can alter the functional state of the tissue. Another requirement of the d.c. amplifier is the lack of a perceptible z e r o drift and fluctuations therein for 20-30 min. The amplification factor should be sufficient to permit recording of changes in potential of the order of 0.1 mV. The circuit of a d.c. amplifier possessing stability and the lack of drift is described in Chapter 11. The frequency characteristic is uniform in a range from 0-2000 c/s. Most of our investigations were conducted with an amplifier of this kind. The apparatus should permit recording a slow speed (6 cm/min). An electrocorticogram can be recorded simultaneously with the ISPO from the same electrodes using a low-frequency amplifier of the kind ordinarily employed to record biopotentials.

OSCILLATIONS IN THE HEMISPHERES OF A WAKING ANIMAL

41

Infraslow rhythms on the surface of the cerebral cortex of the rabbit This investigation contains the data obtained from more than 140 rabbits with chronically implanted electrodes and from 80 rabbits in acute experiments. The experimental animals were not anesthetized. In the acute experiments they were immobilized with diplacin (a curare-like agent) and given artificial respiration. ISPO's can be recorded in all regions of the cerebral cortex, but they do not necessarily appear at the same time. Differences in frequency and amplitude may be observed in different regions (Fig. 8). For example, a rapid rhythm is characteristic of the frontal region, whereas a slower rhythm is characteristic of the occipital region. Two-channel recording of ISPO from symmetrical a r e a s in both cortices shows that they a r e generally asynchronous (Fig. 9).

-w c

Fig. 8. A-rhythm recorded with different leads from the cortex of the same animal. (A) = frontal region of the right cortex (electrodes 1-2); (B) = frontal region of the left cortex (electrodes 5 - 6 ) ; (C) = occipital region of the right cortex (electrodes 3-4).

*

1 min

*

Fig. 9. Recording of infraslow potential oscillations (ISPO) from symmetrical frontal regions of the cerebral cortex.

I min

I

IlmV

Fig. 10. Examples of ISPO in the cerebral cortex of wakeful rabbits (different experiments).

42

INFRASLOW RHYTHMIC POTENTIAL OSCILLATIONS

According to the data obtained from 120 animals with implanted electrodes, ISPO's were equally frequent in both the right and left cortex, but rhythms were recorded in 185 cases (out of 240) in the frontal r e gion and in 110 cases in the occipital region. The rhythms were imperceptible with this amplification at any of the leads in 22 resting animals, but they could be induced by external influences. Infraslow rhythms are irregular in resting animals and their amplitude is low. Fig. 10 presents examples of infraslow oscillations found in wakeful animals not subjected to any external stimulation.

Infraslow rhythms i n different layers of the cerebral cortex Infraslow rhythmic potential oscillations were investigated in different layers of the sensorimotor cortex by means of nonpolarizable microelectrodes (Aladjalova and Koshtoyants, 1957). Diplacin (2% solution injected intravenously) was used to keep the animals quiet. A microelectrode 8-12 p in diameter was inserted perpendicular to the surface of the cerebral cortex. Both bipolar recording with electrodes placed at different levels of the cortex, and unipolar recording was employed. In the latter case, a small silver disc placed on the symmetrical spot of the other cortex served as the indifferent electrode. Fig. 11 illustrates infraslow rhythms recorded under different A

D

lrnin

1

I ~ V

Fig. 11. Examples of recordings of ISPO's by different electrodes with unipolar and bipolar derivations. (A) = bipolar recording by means of chronically implanted silver disc electrodes from the surface of the sensorimotor region of the cortex; (B) = unipolar recording by microelectrodes from the surface of the cortex (same region); (C) = bipolar recording by microelectrodes; (D) = bipolar derivation from the surface and from a depth of 1.5 mm; (E) = disappearance of infraslow rhythms under ether anesthesia.

conditions of derivation. Record A was made with chronically implanted disc electrodes from the surface of the sensorimotor cortex. Record B was made with a nonpolarizable microelectrode and unipolar derivation from the same part of the cortex. Curves A and B show an A-rhythm with a frequency of 6-7 osc/min. Record C was obtained with bipolar derivation of potential between points on the surface of the cortex and

OSCILLATIONS IN THE HEMISPHERES OF A WAKING ANIMAL

43

at a depth of 0.7 mm. In addition to a rhythm of 7 osc/min, this curve sometimes showed even less frequent waves (4-5 per min). Approximately the same picture (D) was observed with bipolar recording between the surface of the cortex and apoint 1.5 mm deep. It is evident from record E that the infraslow rhythm disappears under deep ether anesthesia. ISPO's at different levels of the cortex differ in frequency and amplitude. For example, a rhythm of about 7 osc/min with an amplitude of 0.9 mV w a s recorded with bipolar derivation from the surface, whereas a rhythm of about 4 osc/min with an amplitude of 1.5 mV was recorded by the same microelectrodes at a depth of 1.5 mm (Fig. 12). The

C

B IlmV lmin I

4

Fig. 12. Microelectrode recording of an infraslow rhythm of different frequency from surface (A) and deep layers (B and C : 1.4 mm) of the cortex (bipolar recording).

frequency in the deep layers of the cortex changed slightly with time, e . g . , from 4 to about 3 osc/min. Successively deeper implantation of a microelectrode (unipolar derivation) into the cortex likewise resulted in the recording of ISPO of different frequencies. For example, a rhythm of 6-7 osc/min was recorded with the microelectrode 0.5 and 1 mm deep; at a depth of 1.5 mm the same A-rhythm was superposed on B-waves, while at a depth of 2 mm only a B-rhythm of 1-2 osc/min was recorded. In order to make s u r e that the difference in rhythm on the surface and in the subjacent layers of the cortex w a s not caused by the procedure followed in shifting (lowering o r raising) the microelectrodes, w e used a method involving three simultaneously implanted nonpolarizable microelectrodes. Fig. 13 shows recordings made in the following experiment. Three microelectrodes were inserted into the sensorimotor region of the cortex in such a way that the top one was placed on the surface, the middle one at a depth of 1.4 mm, and the bottom one at a depth of 2.0 mm. Rhythms were almost absent in records from the top electrode (A), while ISPO (B) with a frequency of 4 osc/min were recorded from the middle and bottom microelectrodes. In some experiments, the ISPO was of the same frequency in all layers of the cortex. In many

44

INFRASLOW RHYTHMIC P O T E N T I A L OSCILLATIONS

Fig. 13. Bipolar recording from the upper (A) and lower (B) layers of the cortex by means of three microelectrodes.

experiments, the frequency of the ISPO was higher in the upper layers of the cortex (7-8 osc/min) than in the lower layers (3-5 osc), These experiments indicate, first, that the processes reflected in the appearance of ISPO a r e local in character and, secondly, that infraslow potentials do not spread any further than about 1 mm. Therefore, in recording with electrodes placed on the surface of the cortex, it is fair to assume that ISPO's of the surface layer are recorded, i . e . those arising in the region of the apical dendrites. The ISPO apparently arise as a result of local electrochemical processes.

The lack of a direct link between cerebral hemodynamics and infraslow oscillations According to Sepp (1928), cardiac systole and diastole a r e reflected in the cerebral blood vessels in the form of a rhythmic pulse. The experiments of Kedrov and Naumenko (1954) confirmed the pulsating nature of the blood flow in the cerebral vessels in a closed skull. The presence of constant pulsating and respiratory oscillations of intracranial pressure is due to rhythmic fluctuations of the venous oufflow from the cranial cavity, on one hand, and to shifting of the fluid from the cranial to the spinal cavity, on the other (Vassilevskii and Naumenko, 1959). Despite the changes in the volume of the arteries and veins of the brain, the flow of blood in the capillaries remains uniform. Since the brain is enclosed in an airtight cavity and separated from the cranial walls by fluid, the pulse oscillations of pressure in the vascular system have l e s s influence on brain tissue than on other p a r t s of the body (Vassilevskii and Naumenko, 1959). Moskalenko and Naumenko (1957) found that the fluid in the cerebral cavities moves rhythmically. They observed three kinds of waves of intracranial pressure: waves synchronous with the pulse, waves synchronous with respiration, and third-order waves, which arise during asphyxia. Infraslow potential changes in the rabbit do not reflect the respiratory and pulse fluctuations of p r e s s u r e because the latter have a much greater frequency in the rabbit than the infraslow rhythms. Comparison of the infraslow oscillations with the third-order rhythms in blood pressure and fluctuations of spinal fluid revealed no direct r e lationship between these phenomena. The Traube-Hering waves have a period of about 8 sec, i.e., they coincide in frequency with second-order infraslow potential oscillations. Although the absence of synchronism in infraslow oscillations in adjacent parts of the cortex, e . g . , 1 mm away,

FACTORS INFLUENCING INFRASLOW POTENTIAL OSCILLATIONS

45

argues against the existence of a direct relationship between the infraslow waves and the above-mentioned fluctuations of vascular tone. A series of experiments was designed to check this relationship. We evoked Traube-Hering waves in blood p r e s s u r e in an acute experiment on rabbits by stimulating the central end of the vagus nerve. P r e s s u r e w a s recorded in the carotid artery. No direct relationship was observed between the changes in blood p r e s s u r e and the potential waves. When Traube-Hering waves appeared in the blood pressure, infraslow oscillations could not be detected in the frontal region of the cortex, and, conversely, when there were distinct infraslow rhythms, third-order waves in blood p r e s s u r e were absent. Thus, there seems to be no direct relationship between these phenomena. Many other experiments have also shown the absence of a mechanical connection between cerebral hemodynamics and the rhythm of infraslow oscillations. For example, the increase in blood pressure that lasts for several minutes after intravenous injection of epinephrine has no effect on the parameters of the ISPO in the cerebral cortex. The ISPO in the cortex tend to reflect unusual prolonged changes in excitability caused by changes in the rate of the metabolic reactions. FACTORS THAT INCREASE OR DECREASE INFRASLOW POTENTIAL OSCILLATIONS IN THE CEREBRAL CORTEX

Depression of infraslow potential oscillations foltowing stimulation of the cortical surface Injuries to the cortex (mechanical, electrical, chemical) cause a depression of the ISPO, which spreads through the cortex simultaneously with the spread of an aperiodic wave of negative potential at a velocity of about 3 mm/min. Restoration of the ISPO occurs later than appearance of rapid electrical activity after depression. The mechanism of spreading depression was examined in detail in Chapter 11. Fig. 14 presents ISPO and EEG recordings in the case of "contralateral depression". Stimulation was supplied through electrodes 1-2 in the right hemisphere, while recording (bipolar) was made by electrodes 5-6 from the symmetrical region of the left hemisphere. Depression of the ISPO may also be caused by mechanical trauma, which is frequently the reason for the absence of an infraslow rhythm in the recordings made in an acute experiment shortly after an operation. The high sensitivity of the ISPO to injury to the cortical surface is evidence of the role played by the superficial layers of the cortex (layers of the apical dendrites) in this phenomenon. According to Van Harreveld (1957), the dendrites swell during the occurrence of an aperiodic negative potential wave. O u r data (Chapter VIII) reveal that the electrical impedance of the cerebral cortex increases 15-20% at the same time. To conclude, the phenomenon of infraslow potential oscillations is

46

INFRASLOW RHYTHMIC POTENTLAL OSCILLATIONS

Fig. 14. Depression of infraslow potential oscillations after stimulation of the cerebral cortex. (A) Recording of the potential of the electrodes 5 - 6 . (A') A continued. (1) Moment of electrical stimulation by electrodes 1-2, (0, 2, 3 , 4) parallel EEG recording. (B) Restoration of A-rhythm 30 rnin after stimulation,

incompatible with marked deviations in the physicochemical conditions in tissues from those which correspond to normal manifestations of vital activity.

Stimu lation of r e c epto YS Brief, adequate stimulation of the receptors does not produce changes in the parameters of the infraslow rhythm in the corresponding region of the cortex. Only sustained and very intense stimulation applied continuously for 20-40 min accelerates oscillations in the cortex. Stimulation of the taste receptor by applying a bitter substance to the tongue for 20 min resulted in an increase of infraslow activity in the sensorimotor region of the cerebral cortex (Fig. 15).

Fig. 15. Effect of'stimulating the taste receptor. (1) ISPO in the sensorimotor cortex before stimulation; (2) 20 min after application of a bitter substance to the tongue.

Prolonged flashing of a 30 W bulb for 30-40 min produced rhythmic potential oscillations with a period of 10-20 s e c and an amplitude of 1.3 mV in the visual area of the cerebral cortex of the contralateral hemisphere. These ISPO's persisted for some time after the light was turned off. It is interesting to compare this finding with the data of Chang (1952a), who observed that four different processes are involved in changing

FACTORS INFLUENCING INFRASLOW POTENTIAL OSCILLATIONS

47

the excitability of the visual analyzer after prolonged illumination of the retina. He included among them the very slow fluctuations of excitability with a period of 10 sec, i.e., with the period of the infraslow rhythm. The relationship between a change in cortical excitability and the phase of infraslow oscillation i s examined in detail in Chapter N. Illumination of the eye with a flashing light may give rise to an aperiodic wave in the visual cortex, a phenomenon described by Goldring and O'Leary (1951a, b) and by Aleksanyan and Demirchoglyan (1955); we also invariably observed it in our experiments. However, the aperiodic potential wave, as pointed out in Chapter 11, is to be distinguished from the ISPO. In some cases, repeated flashing of a light intensified the ISPO not only in the visual cortex but in the sensorimotor areas of both hemispheres (Fig. 16), the frequency of the oscillations increasing from 8

Fig. 16. Appearance of high-frequency ISPO in different regions of the cerebral cortex after, repeated stimulation of the right "eye with rhythmic flashing of a light. (A) sensorimotor cortex of the left hemisphere; (B) sensorimotor cortex of the right hemisphere; (C) visual cortex of the left hemisphere. Recordings a r e presented for each region made before (top) and after (bottom). Frequency of light flashes: 8 per sec,duration of a single application of the stimulation: 1 5 s e c , number of applications: 10 (intervals: 5 min).

to 12 osc/min. This indicated that intensification of the ISPO was of generalized nature, The fact that receptor stimulation failed to produce rapid changes in the infraslow oscillations in the projection a r e a of the cortex suggests that the influences transmitted to the cortex through a specific afferent pathway are not directly associated with the ISPO phenomenon. Appearance of the phenomenon with a prolonged latent period and aftereffect point to the importance of nonspecific influences. These influences may

48

INFRASLOW RHYTHMIC POTENTIAL OSCILLATIONS

well be exerted through collaterals from the afferent pathways reaching the hypothalamic region. It is fair to assume that intensification of the ISPO in response to prolonged stimulation of the receptors is brought about through the socalled activating metabolic system which forms part of the defense reaction (like the response to "stress") of the body (Graschenkov, 1959).

Stimulation of subcortical structures Since the neuronal structures of the hypothalamic region play an important role in the origin of generalized reactions in the cerebral cortex, it was to be expected that stimulation of the hypothalamus would influence the parameters of the ISPO. To verify this assumption, bipolar electrodes 50 p in diameter, 150 p apart, were implanted in some structures of the hypothalamus and brain stem reticular formation (RF). These experiments a r e discussed in Chapter V. Here we shall touch on only a few aspects of the effects observed. Stimulation of the hypothalamus may evoke a variety of reactions in the electrocorticogram depending on the intensity and duration of the stimulation and on its topography. Stimulation of the ventromedial nucleus of the hypothalamus resulted in high-amplitude slow waves with a long latent period appearing in the EEG of the sensorimotor cortex of both hemispheres. The ISPO increased in amplitude simultaneously with the appearance of high-amplitude slow waves in the EEG 40-50 min after the onset of stimulation. At the same time the infraslow oscillations decreased, slowing its frequency from 8 to 5 osc/min and the amplitude increased from 0.3 to 1 mV. Intensification of the ISPO after stimulation of the hypothalamus was fairly persistent, being evident two or more hours after stimulation. Stimulation of certain subcortical structures also alters the physicochemical conditions that affect the functioning of the cortical neurons. This conclusion i s suggested by the results of measuring the conductivity properties of the cerebral cortex after stimulation of the fornix, which has extensive subcortical connections ( c f . Chapter VIII). The physicochemical changes in the upper layers of the cortex after subcortical stimulation occur in various ways, depending on the time elapsing since the stimulation. Intensification of the ISPO in the cortex is apparently related to the nature of the chemical changes in the upper layers of the cortex particularly in the a r e a of the apical dendrites. Stimulation of the nuclei of the thalamus and the reticular formation, unlike stimulation of the hypothalamus, does not result in intensification of the ISPO in the cortex, although it evokes an arousal reaction in the electrocorticogram. Moreover, stimulation of the nonspecific nuclei of the thalamus may even depress the ISPO in the sensorimotor cortex (Aladjalova, 1958b). A comparison of these different effects from stimulation of the hypothalamus, on one hand, and stimulation of the reticular formation, on the other, indicates that the hormonal function of the hypothalamus may have major significance in the phenomenon of the infraslow rhythm.

FACTORS INFLUENCING INFRASLOW POTENTIAL OSCILLATIONS

49

I A

Fig. 1 Appearance of high-amplitude slow waves in the EEG simL,aneously with intensification of ISPO following stimulation of the ventromedial nucleus of the hypothalamus. (I) ISPO before (A) and after (B) stimulation; (11) electrocorticogram: (1)= before stimulation; (2) = after stimulation for 60 s e c ; (3) = after a break of 40 min followed by 1 2 sec of stimulation; (4)= after 4 additional stimulations of 10 sec each; (5) = 6 min after the last stimulation.

Effect obtained by resection of cervical sympathetic ganglion

The significance of the metabolic level of the cerebral cortex, for the appearance of ISPO is revealed by experiments with chronical removal of the cervical sympathetic ganglion. According to Orbeli's theory on the adaptative-trophic role of the sympathetic nervous system, the latter has a tonic effect on the activity of the higher nervous centers. The significance of the superior cervical sympathetic ganglia for cortical function w a s demonstrated by the investigations of Asratyan (1930), Maiorov et a l . (1956), Karamyan (1958, 1959), and others. Their r e sults were ascribed to the influence of the cervical ganglia on the state of cortical trophism. It therefore seemed worthwhile to study the effect of resection of the sympathetic ganglia on the ISPO. After unilateral resection of the super i o r and inferior cervical ganglia, the nature of the ISPO changed in the sensorimotor cortex of both cortices. This was preceded by changes in the character of the EEG. For example, on the 5th day after resection the EEG of both cortices exhibited high-amplitude waves, which were slower on the contralateral side. However, on the side of the resection

50

INFRASLOW RHYTHMIC POTENTIAL OSCILLATIONS

the electrical activity at times acquired the character of high-amplitude discharges. The phenomenon w a s particularly pronounced for a period of 7-10 days, when there were occasional bursts of activity at a frequency of 8 c/s (Fig. 18). On the 15th day, electrical activity in the

Fig. 18. Effect of chronic sympathectomy on ISPO and E E G of the cerebral cortex. (I) left cortex; (11) right cortex; (A) ISPO; (B) electrocorticogram. The figures between the records designate the number of days after left resection of the superior and inferior sympathetic ganglia.

contralateral cortex was virtually normal, although the electrocorticogram on the same side continued to show high-amplitude waves. In the 7-10 days period following removal of the ganglia, when changes in the electrocorticogram were most pronounced, ISPO's of considerable amplitude appeared in both cortices. However, their frequency was about the same as that in intact animals, 7-8 osc/min. Electrical impedance of the cortex increased between the same electrodes, suggesting that sympathectomy influences the physicochemical processes in the cortex, which become maximal in both hemispheres by the 2nd week after removal of the ganglia. The ISPO and EEG of the cortex are likewise highly pronounced in the 2nd week. Marked physicochemical and functional impairment is evidently taking place in the cortex at this time. Since these changes are manifested only after chronic sympathectomy, they may be mediated through impairment of cerebral circulation rather than caused by the direct effect of sympathectomy.

FACTORS INFLUENCING INFRASLOW POTENTIAL OSCILLATIONS

51

Role of certain metabolic shifts

The factors that are most effective in increasing the ISPO are changes in the conditions governing the blood supply and the influence of hormones and pharmacological agents thereon. Asphyxia increases the amplitude and regularity of the ISPO, especially if prolonged o r applied repeatedly. As Fig. 19, A shows, partial

I

1min

#

11mV

Fig. 19. Infraslow rhythms in the cerebral cortex during asphyxia; (A) regular infraslow rhythms with partial but not prolonged interruption of respiration. The arrows designate the onset and end of partial compression of the trachea (from left to right); (B) intensification of the A-rhythm due to repeated asphyxiation; (a) = irregular rhythms in an intact animal, (b) = increased frequency and regularity of the rhythm after several compressions of the trachea.

compression of the trachea gradually intensifies the infraslow rhythm in the cortex. Brief compression of the trachea 4-6 times an hour gives r i s e to ISPO's with a frequency of 7 per min and amplitude of 1.5 mV (Fig. 19,B). The effect of prolonged or repeated actions shows up in different experiments. For example, a single subcutaneous injection of 0.25 mg/kg of strychnine may not increase the amplitude of the ISPO in the cortex, although it does evoke a convulsive movement during which an aperiodic potential wave appears along with depression of the infraslow oscillations. Repeated injections (up to 3 per week) cause a persistent intensification of the ISPO,the frequency of the infraslow rhythm increasing from 7 or 8 to 11 osc/min (Fig. 20). Injection of mercury bichloride (3 mg/kg intravenously) markedly inc r e a s e s the amplitude of the infraslow oscillations 50 rnin later. After 90 min the effect diminishes and by the 185th min the picture of the infraslow rhythms can no longer be distinguished from the normal. Three injections of the drug over 48 h, increases the frequency of the infraslow oscillations (about 14 per min). The mechanism of action of these substances when injected into the peripheral blood stream is p e r haps indirect. Injection of hormones likewise changes the ISPO. For example, a lag in intensification of the ISPO occurs after the injection of cortin, pituitrin, or insulin. Injection of insulin (4 I.U./kg intravenously) intensifies both the infraslow rhythm and EEG. Slow waves of 3-4 c/s appear in the electrocorticogram beginning with the 25th min. At the 40th min alternating s e r i e s of rapid and slow oscillations a r e clearly seen, coinciding with a definite phase of the infraslow wave. The ISPO at this

52

INFRASLOW RHYTHMIC POTENTIAL OSCILLATIONS

I

fmin

Fig. 20. Effect of repeated injections of strychnine on ISPO in the cerebral cortex. (1) = before injection; (2) = after the first subcutaneous injection (0.25 m g / k g ) , the arrow designating the time of onset of convulsions); (3) = after 7 days during which strychnine was injected 3 times.

time become regular. Their intensity markedly diminishes between the 70th and 185th min. The effects of injecting hormonal substances show that the intensified ISPO resulting from a variety of agents may be the consequence of a neuro-endocrinal reaction of the organism. Atropine (10 mg/kg intravenously) causes a slight increase in the reflex excitability of rabbits, although the electrocorticogram shows slow "sleep" waves similar to those described in wakeful cats by Bradley and E k e s (1957). There is a divergence between the changes in electrical activity of the cortex (according to the usual interpretation) and the animal's condition. ISPO double o r triple in amplitude 6-60 min after the injection of atropine, but their frequency diminishes somewhat, from 8 to 6 per min. Fig. 21 shows recordings from the

IlmV

imin

4

Fig. 21. Effect of atropine (intravenous injection of 10 mg/kg) on ISPO in the frontal cortex. (1)=before injection; ( 2 , 3 , 4) = 6 , 30, and 60 rnin after injection, respectively.

frontal region of the right cortex before and at different times after the injection of atropine. Injection of acetylcholine into the carotid artery (0.015 mg/kg) results in an increase of the ISPO in the sensorimotor cortex within 2-5 min.

FACTORS INFLUENCING INFRASLOW POTENTIAL OSCILLATIONS

53

Thus, the effect of intensified ISPO due to the injection of acetylcholine o r atropine i s manifested some time after the action as well as after relatively short intervals of time, about 2-6 min. The fact that ISPO's are intensified by substances with such varied mechanisms of action implies that their influence is nonspecific in character. All the factors have one thing in common, namely, they are injurious to the organism, and the central nervous system reacts to them as to stress. As will be demonstrated on more than one occasion, intensification of the ISPO in the brain invariably accompanies this reaction. Two other details are worth noting: (1) infraslow activity is sometimes intensified only after repeated actions; (2) a long latency period may ensue between the time of action and intensification of the

ISPO.

Effect of sleep-inducing and narcotic substances

-----

Study of the effect of narcotics and sleep inducing drugs on ISPO may throw some light on the mechanism reponsible for the origin of the infraslow rhythms. We tested ether, magnesium, urethane, barbiturates, and sodium bromide. The results a r e shown in Fig. 22.

+&W B

2

3

3

-L lmin

,

I 1mV

Fig. 22. Effect of narcotics on ISPO in the cerebral cortex. (A) ether anesthesia (sensorimotor region): (1) = before administration; (2) = during extinction of the blinking reflex; (3) = with deep anesthesia; (4) = 25 min after removal of the mask; (B) barbiturate narcosis (luminal 70 mg/kg subcutaneously); (1) = sensorimotor a r e a of the left cortex before administration of luminal (A-rhythm); (2) = 35 min after administration; (3) = 110 min after administration; (4) = occipital a r e a of right cortex before administration of luminal (B-rhythm); (5) = 25 min after administration: a = electrocorticogram, b = ISPO (simultaneous recording); (6) = after 2 h ; (C) magnesium narcosis: (I) = before administration (sensorimotor area); (2) = after 110 min; (D) sleep (intraperitoneal injection of 1 g/kg of sodium bromide): (1) = before administration (frontal a r e a of left cortex); (2) = after 12 min; (3) = after 28 min.

Ether induces complete depression of the infraslow rhythms with no correlation between the physiological manifestation of the anesthetic and changes in the parameters of the oscillations. Before administration, a rhythm of 4-5 osc/min with an amplitude of 1-1.5 mV (Al) w a s noted in the sensorimotor a r e a of the right cortex. Within 12 min there was a lowering of reflex excitability followed by disappearance of the blinking reflex. The ISPO were somewhat intensified at this time (A2). However, continued administration of the anesthetic until the absence

54

INFRASLOW RHYTHMIC P O T E N T I A L OSCILLATIONS

of a reaction to pain stimulation resulted in complete disappearance of the infraslow rhythms (A3). They reappeared 20-30 min after the mask was removed (A4). Injection of luminal (70 mg/kg subcutaneously) in the stage of superficial narcosis caused a reduction in the frequency of the ISPO with some increase in the amplitude; in the stage of deep narcosis, however, it resulted in their complete depression. Fig. 22,B shows the rhythms in the sensorimotor a r e a of the left cortex (Bl) and in the occipital area of the right cortex (B4) before the experiment. The f r e quency of the A-rhythm in the sensorimotor zone decreased from 10 to 5 osc/min during the 35th min while the amplitude decreased from 0.9 to 0.3 mV (BZ). The depression persisted even 110 min after injection of the narcotic (B3). The pattern of the rhythm was little altered during the 25th min in the occipital region, although the spindles characteristic of barbiturate action could be seen in the EEG. High-amplitude spindle waves coincided (B5) with a maximum of the infraslow wave (two-channel recording a, b). Depression of the B-rhythm was quite pronounced during the 120th min. Emergence from narcosis resulted in prolonged restoration of the infraslow rhythms. Another barbiturate, pentothal sodium -(750 mg/kg subcutaneously), produced a similar depression of the infraslow rhythms, reflected at f i r s t in a decreased frequency of oscillations and then in a reduction of their amplitude until they disappeared entirely. Urethane (1000 mg/kg intravenously) did not produce a complete depression of the ISPO.Despite the disappearance of reflex excitability and appearance of the characteristic slow waves (2-3 c/s) in the electrocorticogram, there was only a slight decrease in the frequency of the ISPO (from 10 to 7 per min). Magnesium sulfate (1.1 g/kg) injected subcutaneously (25% solution) increased the amplitude of the ISPO (Fig. 22, C) and reduced their frequency from 8 to 4 per min. The effect can be observed for the duration of 120 min. Injection of a sleep producing dose of sodium bromide (1000 mg/kg intraperitoneally) resulted in the infraslow rhythms becoming more regular (D2) in the frontal a r e a of the left cortex @1). There was a decrease in the frequency in the 20th min (D3) from 9 osc/min and an increase in the amplitude from 1 to 4 mV. However, this effect of sodium bromide was not always pronounced. Thus, a uniformly manifested functional state (narcosis) may be accompanied by a variety of shifts in the parameters of the infraslow rhythm. Two phases in the change of ISPO parameters are characteristic of the action of narcotics. In the first, infraslow oscillations decrease in frequency but become more regular and sometimes increase in amplitude. They a r e completely suppressed, however, in the phase of deeper narcosis. The two-phase quality of changes in electrical activity of the cortex under the influence of narcotics is also known from electrocorticographic investigations (Sarkisov, 1935; Robiner, 1954). The period of increased regularity of the ISPO in the initial phase of narcosis may be characterized as the period of superficial narcosis

INFRASLOW POTENTIAL OSCILLATIONS IN THE HUMAN BRAIN

55

and appearance of "spindles" in the electrocorticogram. In this stage, according to Kopylov (1957b), periodicity develops in cortical assimilation of the rhythm of receptor stimulation. This periodicity evidently varies with oscillations of the general level of excitability. Gulyayev (1956), using arrhythmic stimulation, likewise showed that wavelike changes in excitability arise with the onset of sleep. This phenomenon, as will be demonstrated in Chapter IV, may be due to the infraslow rhythms. Of particular interest is the increased amplitude of the ISPO during sleep caused by sodium bromide. Such increase is apparently connected with the processes involved in replenishing energy in the cortical neurons.

INFRASLOW POTENTIAL OSCILLATIONS IN THE HUMAN BRAIN

The above-described patterns of changes in the ISPO led u s to believe that these infraslow electrical processes reflect the general characteristics of functional activity of the nervous system. We became of the opinion that under certain circumstances infraslow rhythmic activity could also be detected in man. We carried out an investigation jointly with Prof. F.V. Bassin in the Institute of Neurology, USSR Academy of Medical Sciences. The derivation of infraslow potentials from the intact human skull is complicated by the fact that changes in the skin potentials a r e likewise of very slow character. Therefore, the first series of experiments was designed to a s s u r e u s of the practical possibility of recording infraslow oscillations of brain potentials from the skin and of distinguishing them from the variety of bioelectrical processes occurring outside of the c e r ebrum. The experiments were performed on persons who had defects in cranial bones resulting mostly from craniocerebral traumas and whose skin had completely regenerated over the defect a r e a at the time of the investigation. A total of 5 such patients were examined. Four different electrode arrangements were used: (1) two electrodes over the area of the bone defect 2-3 cm apart; (2) two electrodes in an a r e a symmetrical to the defect or on another p a r t of the head over an intact area; (3) two electrodes on the mastoid processes; (4) one over the bone defect, the other on the auricle. After testing the degree of polarization of different materials (carbon, zinc, silver, etc.), we selected round plates 15 mm in diameter made of chlorinated silver to serve as the electrodes. Filter paper moistened with physiological solution was placed under the electrodes. We used an amplifier with an vibrating switch at the input so that a change in the stable difference in potentials was marked by a change in the amplitude of the c a r r i e r frequency of 100 c/s ( c f . Chapter 11, section "Recording of Slow Potentials"). We succeeded in recording ISPO in 3 patients (out of the 5 who had skull defects). Ouf. most interesting finding was that the oscillations were much more pronounced when derived from the bone defect a r e a than from any other place. This suggests that the presence of bone under an electrode significantly diminishes the amplitude of the infraslow potentials recorded. ,

56

INFRASLOW RHYTHMIC POTENTIAL OSCILLATIONS

An illustrative recording is shown in Fig, 23.

C

Fig. 23. Infraslow rhythmic potential changes recorded from an area of a skull defect. (A) bipolar recording from the parietal area of the right hemisphere, over an intact skull; (B and C ) bipolar recording from a symmetrical region of the left hemisphere, over a bone defect 3 x 3 cm in size. The recording of the respiratory movement is shown under each curve.

Curve A was recorded from the parietid area with cranial bone p r e s ent under the electrodes. The spikes on the curve were caused by deglutitory movements. Under this curve and the two curves presented below is a recording of the respiratory rhythm. Curves B and C were recorded in the same patient and in the same experiment from the parietal region of the other hemisphere under the area of the bone defect left after removal of a brain tumor a year before the examination. These curves clearly show an ISPO with a rhythm of about 4 per min and amplitude of 0.8-1 mV. The frequent oscillations superposed on the recording were caused by pulsation of the cerebral blood vessels. We observed during the experiment that the rhythm of these frequent oscillations coincided with the heart beat. ISPO's were also recorded in the same patient with only one of the electrodes placed over the bone defect. Essentially only skin potentials were recorded when both electrodes were placed over the mastoid processes. These potentials changed aperiodically, i.e., they changed in response to external stimulation (e.g., after an unexpected question was asked or an unfamiliar person appeared ) in one direction or for a considerable period of time. It is known from the literature that changes in skin potentials normally a r e irregular in character. This s e r i e s of experiments was performed, as mentioned above, on 4 other patients with injured skull. Two of them were also observed to have infraslow oscillations that were more pronounced in the region of the brain over the skin defect, but these rhythms could not be detected in the other two subjects regardless of the method of recording used. A second s e r i e s of experiments tested different types of bipolar and unipolar recording of ISPO in healthy persons. They had no ISPO,but while examining them, we found that unexpected situations (unrelated to the examination) to which the subjects were subjected by pathogenic

INFRASLOW POTENTIAL OSCILLATIONS IN THE HUMAN BRAIN

57

agents promoted the appearance of ISPO. Here a r e two examples. Fig. 24 presents the ISPO of the author of this book obtained by three methods of derivation (a) left frontal region, left mastoid process; (b) right frontal region, right mastoid process; and ( c ) right-left frontal regions. Several investigations were made with these electrode arrangements over a period of a year, but the infraslow rhythms were

A

I

1rnin

I

IlmV

Fig. 24. Intensification of ISPO after severe nervous strain. (A) recording under resting conditions (right frontal region), unipolar derivation; (B, C, D) recording 15 min after nervous strain; (B) right frontal region; (C) left frontal region, unipolar derivation; (D) electrodes placed on left and right frontal regions; (E) 48 h after strain, unipolar derivation from the left frontal region.

not clearly defined (Fig. 24, A). However, on a nade the day after prolonged agitation and intense nervous strain, large B-waves (1.5 osc/min) appeared at all the leads with an amplitude that sometimes reached 2 mV along with A-waves (4-7 oscillations) with an amplitude of 0.81.2 mV (Fig. 24, B, C, D). This infraslow rhythm, apparently evoked by nervous tension, was still present 24 h later (E). The other case involved subject A, a 22 year old healthy male. The potentials were recorded three times over a period of a year, but no clear-cut picture of the infraslow oscillations could be obtained (Fig. 25). The evening before the fourth examination the subject was quite intoxicated. Marked impairment of cardiac activity was noted on the recording day. The recording showed distinct ISPO at all the leads. Relatively low-amplitude infraslow oscillations of 8 per min (frontoparietal recording from the left side) could be seen at first, but then oscillations of 4-5 and 8 per min appeared 0). With derivation from the frontal region of the right hemisphere, we recorded exceptionally clear and persistent oscillations with a rhythm of about 5 per min and an amplitude of about 1 mV (E). The infraslow oscillations sharply decreased in amplitude 3 days after intoxication and eventually disappeared altogether (F, G). But when the same subject became drunk again 3 months later and was examined soon thereafter, the picture of the

58

INFRASLOW RHYTHMIC PO TEN TIA L OSCILLATIONS

E

r

k

1min

'

1fmv

Fig. 25. Increase of ISPO of the human brain after alcohol intoxication. (A) before the experiment, frontoparietal recosding, left hemisphere; (€3, C , D, E) 12 h after intoxication; (B) frontoparietal derivation, left hemisphere; (C) frontal region of left hemisphere, unipolar derivation; (D) recording from the same region, 5 min later; (E) frontal region of the right hemisphere, unipolar derivation; (F and G) recordings made after 3 days, unipolar derivation; ( F ) left frontal region; (G) right frontal region,

potentials proved to be similar to that just described, i.e., it confirmed the fact that intoxication has an effect on the magnitude of the infraslow frequencies. Infraslow rhythms were also recorded in a healthy 24 year old female and less distincly in sow-2 of the other persons examined. We concluded from our findings that under certain conditions infraslow electrical rhythms (A-rhythms, 8 osc/min, B-rhythms, 1-1.5 per min, "intermediate" rhythms, 4-5 per min) can be recorded from the surface of the human head. These rhythms closely resemble those observed in animals. Increase of the ISPO under the influence of factors that produce a "stress reaction'' suggests a possible connection between these oscillations and activation of the hypothalamo-hypophyseal mechanisms, which likewise resembles in many respects the results obtained in animals. Recording the ISPO in the human cerebral cortex may have some implications for clinical and physiological research. The main problem, however, is to determine the factors on which these oscillations depend and to discover the reasons why they do not appear except under certain circumstances.

CONC LU S O N S

We conclude from the data that the cerebral cortex is characterized by infraslow electrical activity manifested in the form of rhythmic potential oscillations. The frequency of the oscillations in recordings from

CONCLUSIONS

59

different regions of the cortex and from different depths ranges from 0.5 to 8 per min. ISPO can be recorded from the entire surface of the hemispheres, but they are not necessarily synchronous or isorhythmic. Infraslow activity is characteristic of the brain of a wakeful animal, but it cannot always be recorded distinctly. The regularity of the ISPO increases as a result of more or less systematic influences on the organism to which the central nervous system responds with a rather stable reconstruction of the level of activity. Changes in nervous tissue metabolism within certain limits a r e other factors that increase the regularity of the ISPO. The appearance of infraslow activity in the cerebral cortex is subject to very definite laws. It is inhibited by deep narcosis, and trauma of the surface of the hemispheres. Infraslow activity is intensified by prolonged or repeated stimulation of the receptors or by influences operating on various links in cortical metabolism, e .g., by introduction of pharmacological agents, stimulation of the hypothalamic area, or administration of hormones. Another feature of the ISPO is that they become intensified by agents that elicit a defense reaction similar to the response to "stress". Under these conditions infraslow activity can be recorded in the human brain. A long latency period before the ISPO become intensified, the generalized nature of this intensification, and the prolonged aftereffect testify to the possible participation of special nonspecific brain structures in the phenomenon of infraslow cortical activity. Infraslow activity in the cortex changes in accordance with other temporal parameters than those affecting the more rapid electrical activity. For example, stimulation of the receptors evokes a virtually immediate response in the EEG of the projection area of the cortex, but the nature of the ISPO at this time remains unaltered. Only prolonged or repeated stimulation of the receptors causes intensification of the ISPO after 20-40 min both in the projection area and in other parts of the cortex. The fast reactions of the organism to a single stimulation (possibly accidental) of the external environment are effected by a different s y s tem than that governing the slow reactions, which a r e manifested in intensification of infraslow activity, The possibility of separate functioning of the rapid and slow regulatory systems is confirmed by the difference in the effects of stimulation of the brain stem reticular formation and ventromedial hypothalamus. A single stimulation of the reticular formation immediately elicits an arousal. reaction in the EEG of the cortex, but has no effect on infraslow activity. This reaction is apparently regulated by the rapid regulatory system. Stimulation of the ventromedid part of the hypothalamus several times intensifies infraslow cortical activity within 30-40 min. This reaction is presumably regulated by the slow regulatory system. The slower acting regulatory system may influence the function of the faster acting system, for example, by regulating the exeitable properties of the cortex caused by influences on the local chemical gradients in the neuron populations. Some mechanisms governing the interaction of the slow and rapid activities will be examined in the next chapter.

60 CHAPTER IV

EXCITABILITY OF NEURONS IN THE CEREBRAL CORTEX AND INFRASLOW POTENTIAL OSCILLATIONS

INFRASLOW POTENTIAL OSCILLATIONS COMPARED WITH THE E LE CTROCORTICOGRAM

Infraslow rhythmic potential oscillations (ISPO) directly reflect other processes than does the more rapid electrical activity, although they are interlinked with the latter. An electrocorticogram derived from the surface of the cerebral hemispheres is made up of potential oscillations that reflect the nonlinear interaction of several sources of electromotive force located in a volume conductor. Therefore, it is rather difficult to isolate any of the components of cellular activity in such a complex ECoG. Presumably, the spike potentials of single cells do not reach the surface of the cortex directly because they become extinguished close to the source due to the frequency characteristic of the volume conductor (Chapter VIII). On the other hand, the relatively slow potential oscillations spread with less weakening. Accordingly, there a r e grounds for believing that the postsynaptic potentials, which reflect the slower processes (particularly the dendrite potentials that are generated very close to the electrodes), may be reflected in the electrocorticogram. The appearance of slow high-amplitude waves in an electrocorticogram is regarded by Rusinov (1954)as a "synchronization" of local slow potentials. Such "synchronized" oscillations in the electrocorticogram are most likely to appear when the animal's level of wakefulness is lowered. Extreme synchronization may also occur in pathological states. On the other hand, disappearance of the slower oscillations while the more rapid ones p e r s i s t is characteristic of the reaction of a wakeful animal and its excited state. The presence of "synchronized" oscillations by no means signifies that all the neurons have the same function. Jasper et aZ. (1958) showed that the neighboring neurons may be in opposite states at this time, excited and inhibited. A system consisting of many oscillators interlinked by nonlinear circuits may develop a new rhythm that is not characteristic of any of the elements composing the system. It is quite possible that some oscillatory constituents may arise in the ECoG precisely in accordance with this principle. However, despite the lack of complete certainty regarding the origin of the rhythmic oscillations, the available electroencephalographic data indicate that the type of oscillations in an electrocorticogram is of great significance in clarifying the functional state of the divisions of the brain. The nature of the ECoG sometimes chapges regularly in time with the rhythm of the infraslow potentials. This is shown by the change in degree of synchronization of the oscillations in the ECoG at a maximum and at a minimum of the infraslow wave. Then if bursts of activity

INFRASLOW POTENTIAL OSCILLATIONS COMPARED WITH ECoG

61

appear in the ECoG in the'form of high-amplitude oscillations, they may arise during a definite phase of the infraslow wave and recur with the ISPO period. During "convulsive" activity a perodic shift in the s e r i e s of oscillations with different frequency likewise coincides with the transition of one ISPO phase to another. Typical examples of rabbit ECoG's a r e presented in Fig. 26.

8

Fig. 26. Changes in rabbit electrocorticogram during different phases of slow potential oscillation. (A) sensorimotor cortex (recording by an a.c. amplifier with a large time constant); (B) ECoG of the visual cortex at a maximum (left) and a minimum (right) of an infraslow wave; above, recording of ISPO, below, ECoG; ( C ) change in nature of lkonvulsiveltactivity (intravenous injection of 15 mg/kg body weight) of cardiazol in time with the B-rhythm of the ISPO; twochannel recording from the same electrodes with an a.c. (above) and d.c. (below) amplifier; (D) another example of tlconvulsivettoscillations at a maximum (left) and a minimum (right) of the infraslow wave; above, recording with an a s . amplifier, below, with a d.c. amplifier.

Record A shows an increase in amplitude and decrease in frequency of the oscillations in an ECoG of the frontal area at a maximum of the infraslow wave and a decrease in amplitude at a minimum (during light magnesium narcosis). The electrodes were placed in such a way that the more caudal of the two at the time of a maximum was positive relative to the more r o s t r a l electrode. Fig. 26 presents two ECoG's recorded at a maximum and a minimum of an infraslow wave in the visual cortex in a wakeful animal. Synchronized oscillations with a frequency of 5 c/s can be seen at the time of a maximum (left), marked desynchronization of these oscillations at a minimum. The upper recordings were obtained with a d.c. amplifier and they reflect without distortion the change in amplitude of the ISPO.

62

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

The lower recordings were obtained with an a.c. amplifier with a large time constant. The presence of a capacitative coupling in the amplifier distorted in the infraslow component of the record, decreased the amplitude of the ISPO, and displaced the phase. However, the rapid activity, which is transmitted without distortion, appears more distinctly on the lower record due to the substantial degree of amplification. There is a relationship between the ECoG and ISPO phase even after the administration of certain pharmacological agents that cause "convulsive" electrical activity and associated periodic change in the nature of the oscillations in the ECoG (curves C, D). For example, intravenous injection of cardiazol (15 mg/kg) resulted in high-frequency discharges and high-amplitude slow waves alternating in the ECoG. In many experimental conditions, when periodic activity appears in the ECoG, the periods of this activity coincide with a period of infraslow rhythm (Fig. 27). For example, high-amplitude discharges with a

Fig. 27. Periodic electrical activity and ISPO in the sensorimotor cortex due to stimulation of structures in the diencephalon and mesencephalon. (A) after stimulation of the preoptic nucleus of the hypothalamus: (B) after stimulation of the ventromedial nucleus of the hypothalamus; (C) after repeated stimulation of the reticular formation of the mesencephalon (upper record with a d.c.amplifier, lower record with an a.c. amplifier): (D) during rhythmic stimulation of the intralaminar nuclei of the thalamus (upper recording with an a.c. amplifier, lower recording with a d.c. amplifier),

frequency of about 1 c/s appeared in an ECoG of the sensorimotor cortex after 15 min of stimulation (with square pulses at a frequency of 70 c/s, duration of 10 ysec, and amplitude of 4 V) of the preoptic nucleus of the hypothalamus. After 65 min of stimulation these high-amplitude waves began to cluster together in series that coincided with a particular phase of the infraslow oscillations (Fig. 27, A). The latter also became intensified at this time. The effect was most pronounced 120150 min after the start of stimulation. Repeated stimulation of the ventromedial nucleus of the hypothalamus resulted in the appearance of high-amplitude slow waves in the ECoG of the sensorimotor cortex after 120 min; their pattern changed in time with the infraslow rhythm (Fig. 27, B).

INFRASLOW POTENTIAL OSCILLATIONS COMPARED WITH E COG

63

Repeated stimulation of the brain stem reticular formation could lead to periodicity developing in the nature of cortical activity, to the slow synchronized oscillations giving way to the more rapid oscillations. These periods coincide with the rhythm of the infraslow W&ves. Fig. 27, C presents recordings of the same electrodes. In many recordings bursts of electrical activity did not occur at the very c r e s t of the infraslow wave, but coincided with a r i s e or f a l l in the ISPO record. In other words, the rapid electrical activity occurs during the phase of the steepest changes in the infraslow potential gradient. Electrical activity in the cortex sometimes undergoes periodic weakening and intensification. The periods of these changes usually last 5-10 sec, i.e., they resemble the period of the A-rhythm of the ISPO. Similar periodic changes characterize the electrical reaction of the cerebral cortex to rhythmic stimulation of the sciatic nerve (Moruzzi et al., 1950) and to stimulation of the lateral geniculate body (Chang, 1952a). The period of weakening and intensification of the r e cruiting potentials arising in the cortex in response to stimulation of the intralaminar thalamic nuclei likewise lasts 7-8 sec (Morison and Dempsey, 1943). Simultaneous recording of the infraslow potential (Fig. 27, D) shows that a change in magnitude of the recruiting potential may coincide with the rhythm of the ISPO. It will be noted that visual analysis of the ECoG does not always r e veal a correlation between the ISPO and nature of the rapid activity. Quite often, despite the presence of regular slow waves, the pattern of the more rapid electrical activity is steady. And, contrariwise, spindles can be seen in the ECoG despite the absence of a distinct ISPO. It is possible that the processes responsible for the correlation between the ISPO and the ECoG are not always evident because other factors may predominate. Even though the electrodes a r e bipolar, bursts of electrical activity can be seen, as as rule, only at one extremum of the ISPO.This means that there is no functional symmetry with regard to the middle level of the infraslow oscillation. A case where the potential at electrode A is higher than that at electrode B (ISPO maximum) is not equivalent to a case where the potential at electrode B is higher than that at electrode A (ISPO minimum). Consequently, the direction from A to B is different from the direction from B to A as far as-the relationship between the sign of the infraslow potential and the shape of the ECoG is concerned. There is a change of about 20-30 m V in the difference in the "quasisteady'' potential. The amplitude of the ISPO constitutes a small part of this steady level so that it is conceivable that the change in amplitude of the rapid activity at an ISPO maximum and minimum is not due to change in level of the d.c. potential. All this indicates the presence of asymmetry, which is specific for the oscillatory constituent of the potential. The general level of the "quasi-steady" potential between two electrodes has another value than the difference in potentials, which fluctuates with the infraslow rhythm. According to Caspers (1959), the level of the "quasi-steady" potential reflects the degree of depolarization of the dendrites and is thus an indicator of the level of excitation. In this case slight additional depolarization or hyperpolarization of the

64

EXCITABILITY OF NEURONS IN THE CEREBRAL CORTEX

dendrites could influence their activity. Besides the relationship that is sometimes manifested between the ISPO phase and the ECoG, there is also a relationship between the degree of "synchronization" of potential oscillations in the ECoG and the increase in regularity and amplitude of the infraslow waves. Many fact o r s that increase the amplitude of the ISPO bring about "synchronized" oscillations in the ECoG, e.g., some pharmacological agents and stimulation of certain subcortical structures. Thus, strychnine intravenously injected in a dose somewhat below the convulsive dose causes highamplitude slow waves. These ECoG changes a r e paralleled by an increase in the amplitude of the infraslow waves, which acquire a more regular shape similar t o the sinusoidal while decreasing in frequency from 8 to 6 osc/min. Comparable correlations a r e also found between the origin of synchronous oscillations in the ECoG and intensification of the ISPO after the action of mercury bichloride, insulin, and other substances mentioned in the preceding chapter. Stimulation of the ventromedial nucleus of the hypothalamus (Fig. 17) may likewise result in the appearance of high-amplitude slow waves in the ECoG along with increased amplitude and regularity of the ISPO. The infraslow waves in different regions of both hemispheres become synchronous after stimulation of the hypothalamus. Intensification of the ISPO in these cases i s the result of a generalized effect produced by the subcortical centers whose influence is probably mediated by the endocrine system. The relationship between the nature of the oscillations in the ECoG and the phase of the infraslow wave is apparently based on periodic changes in excitability of cortical elements.

CHANGES IN CORTICAL REACTIVITY ASSOCIATED WITH INFRASLOW POTENTIAL OSCILLATIONS

A change in the nature of electrical activity at the c r e s t s of an infraslow wave sometimes indicates a possible connection between neuron activity and ISPO. Further analysis of this relationship made it necessary to determine which neurons o r which neuron elements are the connecting links. This was done by studying the parameters of the electrical responses of the cortex to excitation from different sources. An electrical response recorded from the surface of the cortex consists of several components related to excitation of different cortical elements and capable of changing independently. As will be shown below, some of the components change in relation to the phase of the infraslow wave. We analyzed potentials derived from the cortex following stimulation of a specific sensory pathway, recruiting potentials and superficial responses to direct stimulation of the cortex.

CHANGES IN CORTICAL REACTIVITY

65

Rhythmic Stimulation of the visual receptor

Livanov (1944a, b) suggested a method of analyzing the instability* of the visual analyzer from the response of the visual cortex to intermittent flashing of a light into the eye. Three aspects of this indicator were studied: (a) ability of the visual cortex to assimilate the rhythm of the stimulus, i.e., to respond with an electrical response to each light flash; (b) threshold of appearance of this response; (c) persistence of the assimilated rhythm some time after the light is turned off. These aspects of cortical activity have become very popular in recent years both in physiological and in clinical investigations (Kopylov, 1957a, b). The ability of.the cortical p a r t of the analyzer to assimilate the rhythm of a stimulus is impaired when cortical excitability is lowered (Smirnov, '1 953). The phenomenon of rhythm "deceleration" o r "transformation" arises when the cortex does not respond with an electrical reaction to each stimulus. Another indicator of the state of analyzer excitability is the length of the latency period of the response to each stimulus (Krigel and Nestianu, 1958). The latency period of the reaction to prolonged intermittent stimulation is marked by regular periodic oscillations with a period of about 300 msec. Chang (1952a) described rhythmic changes in cat visual analyzer excitability with a period of about 10 s e c following prolonged illumination of the retina. An indicator of excitability was the electrical response of the visual cortex (periodically changing in amplitude) to stimulation of the lateral geniculate body with a frequency of 3.5 c/s. This periodicity of changes in visual cortex excitability occurred in our experiments too. We found that prolonged illumination of the retina induced changes in the ISPO, the periodicity being the same as the periodicity of changes in excitability of the visual cortex (Aladjalova, 1956b; Aladjalova, 1957). The picture of assimilation of the rhythm of intermittent light stimulation also changes in accordance with the infraslow rhythm. Electrical activity of the cortex at the start of retinal illumination is quite pronounced (Fig. 28, A) in response to each stimulus. After 30-40 min of stimulation, the cortical reaction becomes periodic, periods of rhythm assimilation, but with a polyphasic form of response (B), are succeeded by periods of rhythm "deceleration" o r "transformation" (C). Fig. 28, D shows the time of transition from deceleration to rhythm assimilation. Simultaneous recording of the ISPO reveals that the period of rhythm assimilation coincides with an inc r e a s e in positive potential on the surface of the visual area relative to the parietal a r e a of the cortex. A change in potential of the visual area in a negative direction is paralleled by a period of rhythm transformation. The phenomena observed are evidently caused by interlinked changes in potential of the deep layers of the cortex. Thus, cortical reactivity may change with the rhythm of the ISPO. Analysis of the origin of the phases of the primary response (Roitbak, *Vvedenskii's concept of instability is applied here to neuron complexes rather than to single neurons.

66

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

Fig. 28. Electrical activity curve of the visual cortex after prolonged flashing of a light at a frequency of 8 per sec. (A) start of flashings; (1)= EEG; (2) = "quasi-steadytf potential; (3) = mark of flashings; (B) reaction following the flashings in the period of a minimum of the B-rhythm (increase in the positive potential of the visual area) ; (C) "transformationtt of rhythm (augmentation of negative potential in the visual area); (D) time of change in EEG corresponding to transition of the infraslow potential from a maximum to a minimum.

1955; Li et al., 1956a) suggests that: (1)the initial positive potential on the surface a r i s e s from excitation of elements in cortical layers 111 and IV;(2) the succeeding negative potential of the primary response i s due to excitation of the dendrites. It has been found that the duration and amplitude of the superficial negative phase increase in the "deceleration" period of the rhythm of cortical responses in parallel with the intensified activity of the dendrites. This intensification occurs at one extremum of the infraslow wave while the dendrite potentials decrease at another. Simultaneous measurement of electrical impedance in the upper layers of the cortex shows that impedance in the dendrite area decreases in the "deceleration" period but increases in the "assimilation" period. This suggests that the quantity of free ions increases (cf. Chapter VII) during "deceleration" due to intensified depolarization of the dendrites in the medium surrounding them. The process of intensification of dendrite activity may be closely associated with the decrease in excitability of the neuron bodies that occurs when assimilation of the rhythm of cortical stimulation is impaired. On the other hand, an increase in impedance of the upper layers corresponds to the period of regular assimilation of the rhythm and adequate excitability. It is possible that some of the dendrites become hyperpolarized at this time owing to the decrease in number of f r e e ions in the medium. The pattern, therefore, is as follows: (a) rhythm assimilation is paralleled by an increase in cortical impedance, weakening of the surface

CHANGES IN CORTICAL REACTIVITY

67

negative responses, and change in cortical potential in a positive direction; (b) rhythm "deceleration" is paralleled by a decrease in cortical impedance, intensification of the surface negative responses, and change in infraslow potential in a negative direction.

Direct stimulation of the cortical surface: dendrite potentials Direct stimulation of the cortical surface produces a biphasic, positivenegative potential. A long (10-20 msec) negative potential, which can be recorded 5-10 mm from the point of stimulation, was described by Adrian (1936) and interpreted by him as a response of the superficial structures (dendrites), unlike that from the "depths", which is recorded on the surface as a positive potential. A dendrite potential can be preceded by brief discharges and a small positive potential (Bishop and Clare, 1952; Roitbak, 1955; Grundfest, 1956). The initial positive phase may be absent with weak or infrequent stimulation. It precedes the dendrite potential and is produced, as Roitbak assumes (1955), by postsynaptic stimulation of cells in layers III and IV. The dendrite response is followed by another positive phase which reflects excitation of the neurons in the deeper layers (V and VI). This causes secondary recruitment of the apical dendrites, as shown by the appearance of a negative afterpotential. Roitbak (1955), Grundfest and Purpura (1956), and others analyzed in detail the origin of components of the surface response to direct electrical stimulation of the cortex. Fig. 29 presents electrocorticograms that we recorded jointly with

roopv

A A A A A R A Ah A A A 1 b

A

~

H

~

Fig. 29. Response of rabbit cerebral cortex to direct stimulation. (A and B) recordings of potentials with bipolar electrodes: top, with a d.c. amplifier; bottom, with an a.c. amplifier, Time marking, 2 msec.

Koshtoyants from the surface of a rabbit cortex after direct stimulation. They were derived with two amplifiers (a.c. and d.c.) but from the same bipolar electrodes. The record shows all four elements of the surface response to direct stimulation (after Roitbak).

68

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

22 rabbits were used, immobilized chiefly with diplacin (curare-like agent), for these experiments. Two pairs of electrodes were employed in some of the experiments. Electrodes spaced fairly close together were connected to an a.c. amplifier, while electrodes spaced more widely were connected to a d.c. amplifier. Bipolar as well as unipolar recording was used. Stimulation was caused by pulses lasting 0.2 to 2.5 msec, 4 to 10 V, and frequency of 1 to 10 c/s. A thin plate (grounded through the stimulation circuit) was inserted between the stimulating and recording electrodes to reduce, stimulus artefacts. Another purpose of the experiment was to investigate the mechanisms of the dendrite potential after direct electrical stimulation of the cortex at different extrema of the infraslow waves. The ISPO were measured between the same points on the cortex as those from which the dendrite response was recorded. In some of the experiments, the ISPO were derived from a broader area (space between the electrodes 4-5 mm) while the dendrite potential was being derived from within this area (space between the electrodes 0.5-1.5 mm). The surface response of the cortex differs at an ISPO maximum and minimum, but not in all the components. The small positive potential preceding the negative dendrite potential remains the same at both extrema of the ISPO,but the negative potential is almost double in amplitude and duration. The shift in polarity of the stimulus does not change the shape of the response (Fig. 30).

Fig, 30. Change in form of electrical response of the cerebral cortex surface on direct stimulation at the maximum and minimum of an infraslow wave. (A, B, C, D) level of infraslow potential. (1, 2, 3, 4) corresponding to these levels recordings of the surface response (in recordings 3 and 4 the sign of the stimulation has been reversed). Bottom: superposition of curves recorded at different levels of infraslow potentials. Change of potential in negative direction, upwards. The rabbit was immobilized by diplacin.

Each recording shows an artefact produced by the stimulus lasting 2.5 msec. Record 1 shows the shape of the response at a minimum of the infraslow wave (recording with the same electrodes). The response consists of a positive potential and a negative dendrite potential followed by a positive deflection and negative afterpotential. The initial positive potential lasts about 4 msec, the negative potential 15 msec,

CHANGES IN CORTICAL REACTIVITY

69

the succeeding positive potential 5 msec, and the atterpotential 20 msec. Record 2 (Fig. 30) was made at the time when the level of the infraslow potential between the electrodes 03) changed by 0.5 mV. The initial positive potential in the recording of the response faithfully retains its magnitude and duration. The negative dendrite potential slightly increases in amplitude and lengthens to 20 msec, the succeeding positive potential is absent, and the afterpotential likewise disappears. Record 3 was obtained one second after Record 2, after reversing the sign of the stimulus. Record 4 was the response at an ISPO maximum @), with a change in amplitude of 1 mV. Comparing this record with record 1, there are several important points to note: the invariability of the positive phase, an almost twofold increase in the amplitude of the dendrite potential and a lengthening of it from 15 to 22 msec. Recordings of the response made at different ISPO levels (designated with letters) a r e shown in Fig. 30 (bottom). The above-described changes in amplitude and duration of the dendrite potential at different extrema of the ISPO recurs several times in the same experiment and with the same arrangement of the electrodes, displaying a definite pattern. The pattern is as follows. The dendrite tomponent of the surface response increases at one of the ISPO extrema, while the next phase, which is associated with the manifestation of excitation of neurons in the deeg layers, disappears. The excitability of neurons in the deep layers is evidently decreased and impulses do not reach the layer of the apical dendrites. This agrees with the analysis of the origin of the second positive component of the surface response made by Burns (1951a, b) and Roitbak (1955), and it supports Adrian's hypothesis (1936) that the neurons in the deep layers a r e the source of this positive potential. Fig. 31 reveals the relationship between the ISPO level and the negative afterpotential, apparently reflecting the activity of the neurons in the deep layers. Thus, the amplitude and duration of the electrical response to stimulation of the cortical surface change simultaneously with the infraslow rhythmic change in potential. However, the initial positive potential does not change in time with the ISPO,but the dendrite potential, succeeding positive phase, and negative afterpotential do change. The initial positive potential is resistant to many factors. For example, the administration of d-tubocurarine (3 mg/kg) reduces the dendrite potential considerably, but the preceding positive potential persists (Purpura and Grundfest, 1956). Only large doses of the drug suppress it. Since d-tubocurarine blocks synaptic transmission, the assumption is that the synapses at the dendrites a r e more sensitive to its blocking action than the synapses of the internuncial neurons. The invariability of the initial positive potential at ISPO minima and maxima may likewise reflect the stability of the properties of the neurons in layers III-IV of the cortex. These facts justify the view that the excitability of some cortical neurons changes with the ISPO,while the excitability of other neurons is independent of this phenomenon. The former includes those neurons whose cell body is localized in the lower layers of the cortex and whose

70

EXCITABILITY OF NEURONS IN THE CEREBRAL CORTEX

h h h A A A A A A 4

'25Hz

Fig. 31. Recordings of surface response of the cerebral corttax to direct stimulation at height of the infraslow potential. (A) bipolar recording; (B) unipolar recording: at B several recordings made under identical conditions.

apical dendrites a r e in the upper layers. The latter includes the internuncial neurons of the upper layers. The excitability of the apical dendrites seems to be closely linked with the phenomenon of the infraw slow rhythm. Rhythmic stimulation of the intralaminar nuclei of the thalamus: recruiting potentials

The nuclei of the thalamus are functionally subdivided (Jasper, 1949) into specific and nonspecific. It is assumed that the effect spreads from the nonspecific nuclei to certain cortical areas. Using histological techniques, Adrianov demonstrated several years ago (1958)that the nonspecific nuclei of the thalamus may be associated with a very limited area of the cortex, thus ensuring the localized character of nonspecific influences on the cortex. Local projection to the cortex was found for the dorsomedial and medioventral nuclei of the thalamus; certain nuclei of the midline have no direct connections with the cortex, but a r e connected with other, specific and nonspecific, thalamic nuclei. Stimulation of the nonspecific nuclei changes the cortical response to stimulation of the specific thalami0 nucleus @arikashvili, 1958). Nonspecific impulses may facilitate discharges of units activated by the specific system, both influencing the dendrites and activating other neurons, and thus result in increased circulation of impulses in closed circuits Stimulation of the intralaminar nuclei oi the thalamus produces mono-phasic potentials from the surface of the cortex (Fig. 32) consisting of four main components: (1)a negative potential spreading with decrement (Clare and Bishop, 1956; Li et a l . , 1956b) and regarded as dendrite potential; (2) a positive oscillation reflecting excitation of the subjacent elements and spreading without decrement; (3) a

.

CHANGES IN CORTICAL REACTIVITY

71

Fig. 32. Recruiting potentials in response to stimulation of the dorsomedial nucleus of the thalamus. (I) components of the recruiting potential: (I) = negative (dendrite) component; (2) = positive; (3) = positive trace; (4) = negative trace components. Maximum and minimum: recordings of recruiting potential at the corresponding extrema of the infraslow potential wave. Experiment without anesthesia, bipolar recording from the sensorimotor cortex. (11, A) intensification and weakening of recruiting potentials in the sensorimotor cortex with an 8 sec period; (B) the same, with simultaneous recording by means of a d.c. amplifier (upper curve) and an a.c. amplifier (lower curve); the recordings show the shift from an ISPO minimum to a maximum; ( C , and C,) recordings with greater velocity at an ISPO minimum (left) and maximum (right); (D1 and Dz) recruiting potentials at an ISPO maximum and minimum in an animal with _chronic desympathization of the superior and inferior cervical ganglia.

positive trace wave; (4) a negative trace wave. The surface negative potential is regarded as a reflection of postsynaptic depolarization of the dendrites, which is produced either by direct flow of impulses to the apical dendrites over a nonspecific pathway or indirectly through excitation of neurons deep in the cortex. The mechanism by which the latter influence the dendrite potentials is not known. One hypothesis (Li et al., 195613) is based on the notion of electrotonic spread of postsynaptic potential of the cell body in an ascending direction. Other neurons may be recruited in response to nonspecific stimulation than those responding to impulses over a specific pathway. However, the so-called nonspecific structures influence the conditions under which the neurons associated with a specific function are excited. Substances that inhibit neuron activity (procaine, veratrine, y-aminobutyric acid) change the shape of the recruiting potential complex. The negative potential is inverted into positive, the trace positive wave is increased without a change in polarity, and the trace negative potential is increased (Goldring et al., 1958). However, the shape of the recruiting potential in the deeper layers of the cortex is not changed by these agents. Purpura et al. (1957a, b) ascribe the transformation of the negative

72

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

wave into a positive wave to hyperpolarization of the apical dendrites which becomes dominant due to depression of the exciting (depolarizing) axodendritic synapses and to increased activity of the inhibiting (hyperpolarizing) synapses. In other words, they believe that the positive potential can likewise reflect the response of the apical dendrites. According to Goldring et a l . (1958), the sign of the dendrite potential is inverted as a result of local suppression of the activity of the surface structures while the surface electrode records the positive potential spreading from the deeper layers, i . e . , the sign of the electrical activity is determined by convection of the current and distribution of the potential in the volume conductor. We recorded recruiting potentials in chronic experiments on 8 unanesthetized rabbits. Silver plate electrodes were chronically placed on the sensorimotor cortex 6 mm apart. Bipolar electrodes 30 p in diameter were implanted on the ipsilateral side. The thalamus was stimulated by pulses lasting 10 msec with an amplitude of 4 V and a frequency of '7 c / s . A picture of decrease and increase of the recruiting potentials is created by a change in the appearance of the individual constituents of the potential (Fig. 32). The period of intensification is marked by an increase in amplitude of negative deflection (dendrite potential) and positive and negative trace potentials. The character of the recruiting potentials changes with a period of 7-9 sec, i.e., with the period of the A-rhythm of the ISPO. Simultaneous recording of the ISPO (at the same electrodes) sometimes reveals that the amplitude of the dendrite responses at a maximum of the infraslow wave is almost double than at -a minimum and that the positive trace potential is likewise almost double. Examples are shown in Fig. 32, C, and C,. Fig. 32, D, and D, shows recruiting potentials recorded in a rabbit 7 days after unilateral removal of a cervical sympathetic ganglion. The recordings, made at a maximum and a minimum of the ISPO,reveal that the cortical response at one of the extrema i s distorted. One can see in the record when the initial negative response turned into a positive one. According to this indicator, sympathectomy has the same effect as y-aminobutyric acid. Thus, a factor that intensifies the ISPO due to a change in cortical trophism (in this case sympathectomy, (cf. Chapter IV) also causes a greater difference in the shape of the recruiting potential at a minimum and a maximum of the ISPO. The difference is largely in the dendrite component and in the trace positive phase. Relationship between dendrite potentials and infraslow rhythm

A distinct relationship was noted between the magnitude and latency period of the dendrite response and ISPO phase in all three series of experiments in which dendrite potentials were evoked by various methods of stimulation (direct stimulation of the cortical surface, stimulation of a receptor, aatd stimulation of the nonspecific thalamic nucleus). There were also changes in the trace potentials of the cortical surface. We shall now discuss the possible mechanisms involved in the

CHANGES IN CORTICAL REACTIVITY

73

changes in dendrite potential at infraslow-wave extrema. One hypothesis is that the source of electrogenesis of the ISPO is outside the dendrites. Dendrite potentials a r e postsynaptic in origin and they reflect local excitation. Grundfest (1957a) presents some evidence that the dendrites cannot be excited electrically in a direct way, as is the case with the axons. However, experiments with artificial polarization show that extracellular currents may influence the dendrite potential. It is possible that extracellular currents slightly depolarize o r hyperpolarize the membrane and thus influence dendrite excitability. According to this view, the extracellular currents created by infraslow potential generated in other elements can affect the excitability of the apical dendrites. Another hypothesis is that the infraslow autooscillatory rhythmic processes a r i s e in the dendrites themselves. This possibility is examined in detail in Chapter VI, where it is shown that the ISPO can reflect a rhythmic change in the membrane potential of the nerve f i b e r s and that influences on various elements in metabolism evoke infraslow oscillations of different frequency. The infraslow rhythm is determined by the interaction of active chemical agents (the hormones, in particular) with enzymatic processes of the cell. The development of an auto-oscillatory process on the dendrite membrane may be caused by changes in the excitability of this membrane (depending on the phase of oscillation). Both synaptic and tissue production of hormones may be sources of active substances capable of evoking ISPO in the dendrites. The role of the glia in tissue production of hormones must be taken into consideration (cf. Chapter I). Kornmuller has noted that glia secretion is rhythmic. It has been shown that the neuroglia, at least in the nuclei in the walls of the third ventricle, has a secretory function (Ford and Kantonnis, 1957). The subcortical struct u r e s are said to have neurosecretory fibers that penetrate into the nerve cell to the point where they come into contact with its cytoplasm (Barry, 1954a, b, c, 1956). Microphotographs show drops of secretion that have reached the cytoplasm. The endings of other neurosecretory fibers are similar to those of the axodendritic synapses. The data on rhythmic secretion and pulsation (Chang and Hild, 1959) of the glia (oligodendroglia) cells and their role in neuron metabolism (cf. Chapter I) justify inclusion of the glia among the structures involved in the electrogenesis of the ISPO. This view is supported, for example, by the results of recording ISPO in the area postrema, which is a m a s s of glial cells functioning as chemoreceptors in connection with the activity of the "vomiting" center (Borison and Wang, 1953). ISPO's were recorded in the area postrema with a frequency of 8 osc/min; the parameters of the oscillations changed after local action of the cholinesterase inhibitor. Two hypotheses have been tentatively advanced on the origination site of the ISPO: (1) they a r i s e on the dendrite membrane, (2) they originate in the glia, and secondarily create rhythmic depolarization and hyperpolarization of the dendrite membrane, thereby influencing the excitability of the dendrites. We tend to favor the first hypothesis. The ISPO's are apparently an electrical manifestation of very slow changes taking place in membrane potential in whose origin hormonal factors play an important role. The dendrites are chemically sensitive

74

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

structures rather than electrically excitable ones, a property that is particularly associated with the origination of the auto-oscillatory process therein. The source of the ISPO in the upper layer of the cortex is localized, according to this interpretation, in the dendrite structure. Nor is the interpretation contradicted by the facts on ISPO depression following trauma or injury to the cerebral cortex (Chapter III). Dendrite ISPO’s a r e probably caused by a variety of influences: inflow of excitatory impulses along the axons (from specific and nonspecific structures) and result of the chemical action of the medium. This peculiar function of integration of nervous and humoral influences in the dendrites affects the metabolism of the neuron, its impulse activity, and cyclic changes in its excitability. The electrical field of the infraslow potential may, in turn, be a factor that serves to recruit the excitability of the neighboring neurons into a single rhythm of changes. Such synchronization is facilitated by the fact that the dendrite branches of the neighboring neurons a r e closely intertwined. The extracellular currents created by the source of the infraslow potential (in the dendrites of the particular neuron) may cause periodic depolarization and hyperpolarization of the dendrite potential of the neighboring neuron. INFRASLOW POTENTIAL OSCILLATIONS IN AN ISOLATED STRIP O F CORTEX

Excitation does not reach a strip of cortex that is completely isolated from its surrounding. Thus, if there is no direct (electrical or mechanical) stimulation of this strip to excite the neurons, the processes evoked by synaptic excitation do not appear. Electrical phenomena may arise under these circumstances as a result of chemical factors reaching the cortex, particularly hormones interacting with the chemical structure of the cell. They need not be caused by the occurrence of excitation impulses, and they may simply reflect the slow fluctuations of membrane potential. The method used to isolate a strip or slab of cortex from all its connections (from the adjacent regions of the cortex and from the subcortex) permits broader analysis of the origin of electrical phenomena in the cortex (Burns, 1951a; Burns and Grafstein, 1952). Circulation of blood is fairly well preserved in such a strip because the pia mater is left intact. The sympathetic nerve fibers along the blood vessels a r e also preserved. Burns (1951a, b) showed that in the absence of artificial stimulation there is no spontaneous electrical hctivity in the strip of the isolated cortex, but electrical oscillations arise in response to stimulation of the strip and then become damped after a few minutes. The author concluded that excitation must reached the cortex through the afferent pathways if the so-called “spontaneous” activity is to develop. Several other investigators confirmed the absence of electrical activity in isolated strips of cortex. For example, activity of the cortex of cats, dogs, and man (Echlin et al., 1952) in the absence of anesthesia and with intact

OSCILLATIONS IN AN ISOLATED STRIP O F CORTEX

75

brain stem is characterized by episodic bursts, generally irregular but containing high frequencies of low amplitude (50-70 c/s, 50 WV),and slow waves of large amplitude (2-3.5 c/s, 500 vV); with the strip at r e s t for some time, this spontaneous activity disappears. However, some authors present data on the presence of "spontaneous" activity even in isolated cortex (Kristiansen and Courtois, 1949; Ingvar, 1955b, c ) . Isolated s t r i p s of cortex, even when completely deafferentiated, are capable of generating rhythmic electrical activity. Three causes of this activity are mentioned in the literature (a) direct stimulation of nerve elements of the strip itself, resulting in the appearance of potential oscillations in the form of bursts (Burns, 1954; Echlin et a l . , 1952); (b) chemical stimulation of extraneural nature, influence of the internal environment (Ingvar, 1955a, c ) ; ( c ) presence in the cortex of a natural mechanism that generates specific spontaneous activity, e.g., of the discharge-wave type (Henry and Scoville, 1952). U s e of the "isolated cortex" method may help to answer some of the questions pertaining the origin of the infraslow rhythms: (1) Can ISPO exist independently of rapid "spontaneous" activity ? (2) Is it possible to alter the parameters of the ISPO by stimulating the cortex directly rather than through the subcortical structures? (3) Is there an extraneuronal pathway in the transmission of impulses from the subcortical regions to the ISPO in the cortex?

Methods We performed a s e r i e s of experiments on 52 rabbits with "isolated cortex" jointly with Koshtoyants. The right hemisphere was exposed without the use of anesthesia after preliminary exposure of the bone. ISPO's and electrocorticograms were recorded from the dura mater. We then completely isolated a strip of cortex in the parietal region while preserving blood circulation through the pial vessels. The size of the isolated s t r i p was 6 x 4 mm wide and 3 mm deep. The s t r i p was covered with warm oil and the brain stem was left intact. The animal was then injected with urethane (0.5 g/kg of body weight) o r immobilized with diplacin (curare-like agent) and given artificial respiration. Bipolar silver needle electrodes were used for recording. Wick electrodes were also used in some of the experiments to minimize irritation. The ISPO's were recorded with the aid of a d.c. amplifier. An amplifier with a time constant of 0.5 s e c was used to record the electrocorticogram. The s t r i p of cortex was stimulated with square pulses (50 c/s, 2 V) at a distance of 3 mm from the recording electrodes. Chemical stimulation was supplied by intravenous injection of 0.3 mg) kg of phosphacol (an inhibitor of cholinesterase) and application to the cortex of a sheet of filter paper moistened with 0.1% strychnine solution. Cortical function was checked during the experiment by the response to direct electrical stimulation. Another series of experiments was devoted to investigating the significance of the brain function involved in controlling the activity of the hormonal apparatus for electrical activity of isolated s t r i p s of cortex.

76

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

Bipolar electrodes were implanted in the hypothalamus, chiefly in the preoptic nucleus of the left and right sides of the brain. Plate electrodes were placed on the surface of the right hemisphere. The operation to implant the cortical and subcortical electrodes was performed 7-10 days before the experiment. The strip of cortex in the left hemisphere was prepared on the day of the experiment. Some 30-40 min after the preparation, spontaneous electrical activity and ISPO were r e corded both from the s t r i p and from the symmetrical cortex with the previously implanted electrodes. The hypothalamus was then stimulated on both sides f o r 20 min with square pulses (10 p e c , 6 V, 50 c/s). Electrical activity in the s t r i p and in the intact cortex was observed for 3-4 h afterward. The results of these experiments were compared with the effect of desoxycorticosterone injected intramuscularly (2 mg/kg) and of topical application of serotonin (in a concentration of 0.01%).

Types of electrical activity in isolated strips of cortex Although investigation of spontaneous activity does not f a l l within the scope of this monograph, we shall nevertheless discuss some aspects. "Spontaneous" electrical activity in isolated strips of cortex of the parietal a r e a was observed in 33 cases for a period of time ranging up to 50 min after the strip was prepared. It was not recorded in 18 cases, but it could be induced by electrical stimulation of the strip. "Spontaneous" activity was observed immediately after the strip was prepared, apparently the result of traumatic irritation. Within 20-30 min of the operation "spontaneous" activity either became imperceptible or very irregular. "Spontaneous" activity generdly disappeared some time later, but it could be elicited by stimulating the strip. Fig. 33 shows an example of successive changes in "spontaneous" activity (A) after the s t r i p of cortex was isolated. Within 5 min of isolation (B), electrical activity dropped sharply; the amplitude of oscillations decreased, a low frequency of 1-2 c/s dominating. These oscillations eventually damped out and they appeared only periodically. After 17 min, bursts of high-amplitude oscillations (about 1000 MV)appeared in the ECoG with a frequency of 3 c / s along with a higher-frequency rhythm (about 40 c/s). At first the bursts of activity recurred only infrequently (80 s e c interval); then the intervals between bursts shortened to 7 s e c by the 25th min (C), but later again became infrequent (D) with total electrical "silence" between them. The bursts became irregular by the 45th min and completely disappeared thereafter. Application of a 0.1% solution of strychnine to the cortex adjacent to the isolated portion did not change the electrical activity of the strip. This shows that electrical discharges from adjacent regions are not transmitted to the strip after the anatomical connections were cut. Fig. 33 is an example of "spontaneous" activity of the "burst" type. Similar "spontaneous" activity has been described for isolated cat cortex (Burns, 1952; Ingvar, 1955b). The different kinds of "spontaneous" activity of isolated cortex that we observed are shown in Fig. 34. For example, 25 rnin after isolation (Al),

OSCILLATIONS IN AN ISOLATED STRIP O F CORTEX

Fig. 33. Changes with time in "spontaneous"electrical activity of isolated rabbit cortex. (A) record from the cortical surface before isolation; (B) record from the cortical surface 5 min after isolation; (C) after 25 min; (D) after 35 min; (E) after 45 min. Bipolar recording $ith wick electrodes, without anesthesia.

0

0

w

4

M

x

77

78

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

Fig. 34. Different kinds of ftspontaneowffelectrical activity in isolated rabbit cortex. (Al) 25 min after isolation (urethane aneethesia); (&) 30 min later; (&) electrocardiogram; (B1)16 min after isolation of cortex (no anesthesia); (h) 23 min later; (Ba) 30 min later; (C) 35 rnin after isolation (urethane anesthesia); (Dl) 65 rnin after isolation, at rest (urethane anesthesia); (&) 1 rnin after electrical stimulation of the strip 3 ~ ~ from f m the recording site. Bottom right, diagram of stimulation and recording.

there were regular oscillations with a frequency of 12 c/s; after 30 mjn (4)they appeared only sporadically (urethane anesthesia). On the r i a t is a recording of cardiac activity (4)showing that electrical activity in the strip was not due to vascular pulsation. Al*opgh a rhythm of 1 c/s is equal to the respiratory rhythm, there is no synchronism between the afore-plentioned oscillations and respiratory movements. The ECoG also reveals oscillations that occurred within a frequency of 3 c/s at intervals of 15-25 min after preparation of the strip (BI and B,). They disappeared after 30 min (B3).Finally, Fig. 34 shows the appearance of characteristic high-frequency bursts at the 35th min against a background of slow activity (urethane anesthesia). In 18 cases, "spontaneous" activity in isolated atrip of cortex disappeared 10 min after preparation and later (Fig. 34, Ds),but it could be evoked by electrical, chemical, or mechanicd stimulation. Electrical stimulatian (D,) produced oscillations in two frequency ranges: low, 1-2 CIS with an amplitude of about 500 vV; high, 8-12 c/s with an amplitude of about 100 pV. This electrical activity s o m e t i d s appeared immediately after stimulation, but damped out in 10-30 sec. Hodever, there were also experiments in which "spontaneous" activity damped out and then periodically (every 60-100 sec) intensified without apparent external stimulation. In other cases, "spontaneous" activity did not appear until 50-60 sec after stimulation of the strip and it consisted of pbtential oscillations with a frequency of about 1,4, and 12 c/s. Chemical factors likewise excited "spontaneous" activity in isolated cortex whether applied directly or transmitted humorally. For example (Fig. 35), after the application of strychnine (Q.1%)to isolated cortex, whose activity was marked by highly irregular and weak oscillations

OSCILLATIONS IN A N ISOLATED STRIP O F CORTEX

Fig. 3 5 . Application of strychnine to isolated cortex combined with electrical stimulation (no anesthesia. diplacin}. tAj 6 2 mi11after isolation of cortex: (B) 5 sec after application of strychnine (0.19):( C ) after electrical stiniulation ag,ainst a background of further application of strychnine. 99 min after isolation: tD and E ) c o n h u a tion of C (continuous recordingj: (.F and G) cardiac rhythm at the s t a r t and end of the experiment.

79

80

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

with a frequency of 2-3 c/s (A), high-amplitude discharges with a frequency of 7-12 c/s (B) appeared in the strip. After the strychnine discharges ceased, the strip was stimulated electrically. Electrical activity of the "discharge-wave" type appeared, turning into group discharges (D). At D on the recording, inversion of the potential sign of the first discharge in each group can be seen. The cardiac rhythm r e corded before (F) and after the application of strychnine (G) can be seen at the bottom. It is obvious from this experiment that electrical potentials do not reflect pulsations of the blood vessels, although some of them are unipolar in shape. Thus, "spontaneous" electrical activity in isolated cortex occurred some time after the strip was prepared. In some experiments, this interval w a s only a few minutes long during which electrical activity was characterized by bursts of oscillations with a frequency of 12-40 c/s or by slow waves with a frequency of 1-3 c/s. In other experiments, irregular "spontaneous" activity was observed for an hour after the strip was isolated, but it eventually disappeared. This does not mean that the neurons of the strip lost their ability to be excited because they reacted at this time to electrical stimulation. The nature of the reaction of isolated cortex to stimulation indicates that it is more excitable than intact cortex. Henry and Scoville (1952) believe that inhibitory influences from the subcortical structures are removed by deafferentiation of the cortex and, according to the general principle of increased sensitivity of denervated structures , neuron excitability increases. In addition, excitability of a s t r i p is higher in an animal with an intact brain stem than in a "cerveau isol6" preparation. This suggests that the humoral mechanism may well play a part in realizing the influence of certain subcortical structures on cortical excitability. Stimulation in the region of the dorsomedial hypothalamus can actually cause electrical activity in isolated strips of cortex (Fig. 36).

Fig. 36. Appearance of electrical reaction in isolated cortex to stimulation of the hypothalamus. (A) 100 min after isolation; (B) 10 rnin after stimulation of the medial hypothalamus; ( C ) 64 min later; (D) 167 min later.

Spontaneous activity disappeared 100 rnin after the strip was isolated (A). Stimulation of the hypothalamus (bipolar electrodes, square pulses 10 msec duration, frequency of 50 c/s, 6 V, 5 sec duration) did not cause any immediate changes in the ECoG (B). However, 64 min after stimulation biphasic oscillations appeared periodically in the strip (C), disappearing by the 167th min. The appearance of an effect 50-70 min after stimulation of the

OSCILLATIONS IN AN ISOLATED STRIP OF CORTEX

81

hypothalamus points to the probable existence of a relationship between the effect and release of hormones into the blood. In other words, there is an extraneuronal mechanism of communication between the hypothalamic region and cerebral cortex. Electrical activity has likewise been observed in cat isolated cortex after stimulation of the reticular formation of the brain stem (Ingvar, 1955b). Stimulation of the hypothalamic region results in a twofold change in the electrical activity of isolated cortex: appearance of potential oscillations with a frequency of about 1 CIS 10-15 min after stimulation and oscillations with a frequency of about 3 c/s after 40-60 min. This pattern is illustrated by experiments in which the preoptic nuclei of the hypothalamus were stimulated (Fig. 37). Oscillations with a frequency of about 3 c/s

Fig. 37. Changes in electrical activity of isolated cortex due to stimulation of the left supraoptic (A) and preoptic (B) nuclei of the hypothalamus (urethane anesthesia). (A, 1) ECoG of strip 35 min after isolation; (2) 15 min after stimulation of the supraoptic nucleus; (3) 100 min later. {B, 1) recordings from isolated cortdx; (II) recordings from the symmetrical part of the intact hemisphere; {a, a') = ECoG 65 min after isolation of cortex;-fb, b') = ECoG 15-20 min after the end of stimulation of the hypothalamus; (c) = ECoG of isolated cortex after 30 min; (d) = ECoG of isolated cortex after 75 min; (c') = ECoG of cortex after 120 min.

appeared both in the strip of cortex and in the intact symmetrical p a r t of the cortex. Frequencies of 1-3 c/s appeared in 11 out of 15 experiments involving stimulation of the hypothalamus. Since stimulation of the hypothalamic nuclei causes a variety of reactions (Chapter V), it may be that the appearance of oscillations with frequencies of about 1 and 3 c/s is due to various causes. For example, oscillations at about 1 c/s may be evoked by excitation of the respiratory center, whereas oscillations that appear later with a frequency of about 3 c/s result from a shift in the hormonal level. Thus, "spontaneous" activity in isolated cortex may be sustained by the interaction of chemical agents into the bloodstream. A low-frequency range of activity (1-3 c/s) is associated with the chemical mechanism of stimulation. Beritov (1937) once suggested that the ability of cortical neurons to be excited rhythmically is due to the stimulating effect of chemical substances (hormones, metabolites, electrolytes). Later,

82

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

however, especially after Burns' experiments with "electrical silence" of deafferentiated cortex, the role of chemical humoral factors was relegated to the background. Kristiansen and Courtois' data (1949) on the appearance of electrical oscillations of the "spontaneous" activity type in isolated cortex several hours after isolation were appraised as an artefact caused by the transmission of excitation from the adjacent areas of the cortex @urns, 1951a, b). Ingvar's detailed experiments (1955a, b, c) renewed*interestin the possibility of "spontaneous" activity in isolated cortex. The author observed, for the most part, oscillations with a frequency of 1-2 c/s resembling those shown in Fig. 34. The foregoing considerations suggest that there is a mechanism which generates potential oscillations in isolated strips of cortex after a shift in the hormonal level. This mechanism does not seem to involve excihtion of the synapses but is based on the intimate interaction of hormones with the chemical make-up of the cell that may influence rhythmic fluctuations of membrane potential. Higher frequency rhythmic fluctuations have been described by Phillips (1956b) and Li (1959) for the cortical neuron membrane. The above-described slow electrical phenomena in isolated cortex probably a r i s e as a result of enzymatic processes in cells and are manifested in potential fluctuations of the neuronal membrane. Bursts of more rapid activity that arise only as a result of specific stimulation may be intensified at the crest of a wave of this slower activity. There seem to be two kinds of "spontaneous" activity in isolated cortex. One kind depends on the inflow of impulses of excitation along the nerve fibers; the oscillations a r e of high frequency and they disappear in the absence of cortical stimulation. The other kind is sustained by hormonal factors; the oscillations a r e of low frequency and they a r e caused by the interaction of hormones with the chemical structure of cortical elements. hfraslow rhythmic potential oscillations in isolated straps of cerebral cortex

Infraslow rhythmic potential oscillations may be recorded in isolated strips of cortex. IsPo's were recorded in 46 out of 65 experiments. However, they were absent in 14 cases in intact cortex, but were quite distinct in the isolated strip. In 12 cases, ISPO's were absent both before and after the strip was isolated. They were recorded in 6 cases before isolation, but were absent in the isolated strip. JSPO's did not appear in the isolated strips until 20-50 min after the operation. This may have been due to the effect of acute trauma, which usually disrupts the regularity of the infraslow oscillations and depresses them. For example, in Fig. 38, the recording (A),which was made with chronically implanted electrodes, shows,regular ISPO's with a rhythm of about 7 per min and ampl@de of 1.2 mV. After a trepanation was made in an adjacent part of the skull, the regularity of the ISPO was instantly disrupted (B).Cortex was then isolated in the a r e a under the electrodes. Infraslow rhythms (C) with a frequency of 8 per rnin and an amplitude of about 1.5 mV were recorded 40 min after this operation

OSCILLATIONS IN AN ISOLATED STRIP O F CORTEX

83

A

B

o e -

~

I

1 rnin

-qZT--

Fig. 38. Changes in ISPO in cerebral cortex due to isolation. (A) recording with implanted plate electrodes; (B) the same, after trepanation in the adjacent area of the skull; (C) recording from cortex 40 min after isolation (needle electrodes); (n) after halting of blood circulation through the isolated cortex.

from the isolated strip; the more rapid "spontaneous" activity w a s missing at this time in the strip. Halting of blood circulation in the pia mater (for 2 s e c before recording D) caused the infraslow oscillations to disappear. The ISPO in isolated cortex (Fig. 39) is characterized by rhythms of

Fig. 39. ISPO's in isolated cortex. (A) A-rhythm of 8 osc/min combined with B-rhythm (30 rnin after isolation, urethane); (B) rhythm of 8 osc/min, 16 min after isolation; (C) rhythm of 7 osc/min, 10 min after isolation; (D) B-rhythm of 1osc/min, 10 min after isolation. Biopolar recordings.

approximately the same frequencies, which are found in the area retaining all the nerve connections. The frequency of the rhythms ranges from 8-10 to 0.5 osc/min with an amplitude of 0.5-3 mV, the most common being 5-6 per min. The later distinguishes isolated from intact cortex, in which a rhythm of 8 osc/min is the most common. As in the case of intact cortex, oscillations with two frequencies can be combined in the same recording. The less regular shape of the ISPO in isolated cortex is due to the conditions of an acute experiment. ISPO do not appear until 20-40 min after the strip is isolated, whereas the rapid "spontaneous" activity is most pronounced during the first 10 min after isolation, disappearing by the 60th min. Whereas the "rapid*' spontaneous# activity gradually disappears after the strip of cortex is isolated, the JSPO's, on the other hand, are absent during the first few minutes, but appear later on. There is evidently no immediate direct relationship between the presence of infraslow oscillations and the more rapid "spontaneous" activity. Infraslow rhythms may appear at a time when "spontaneous" activity is either present or absent. The fact that ISPO's are present in isolated cortex indicates that they are generated in the cortical structures. External influences on these

84

EXCITABILITY O F NEURONS IN THE C E R E B U L CORTEX

Fig. 40. Effect of phosphacol and ether on the ISPO of isolated cortex. (I) absence of ISPO in cortex 60 min after isolation (A) and their presence 2 min after injection of phosphacol (B) into the carotid artery (0.25 mg/kg). (11, A) ISPO 41 rnin after isolation of cortex; (B) 55 min later; (C) 5 rnin after intravenous injection of phosphacol (0.3 mg/kg); (D) 35 min later; (E) 50 rnin later; (F) 90 rnin later; (G) 120 min later; (H and r) during ether anesthesia; (J) 40 rnin after removal _of ether mask; (K) 60 min later.

structures would presumably change the parameters of the ISPO. In intact cortex, the ISPO parameters change when the cholinergic neurons are subjected to certain influences (Aladjalova, 1958a). The parameters in isolated cortex change after phosphacol is injected into the bloodstream (12 experiments) in small doses that do not cause convulsions; the amplitude and frequency of the ISPO increase, The effect is manifested at two periods: 2-10 min after injection and 50-100 min later. The appearance of irregular ISPO's 2-10 min (Fig. 40, I) after injection of phosphacol into the carotid artery (0.25 mg/kg) may be due to direct transfer of the cholinesterase inhibitor to the isolated cortex by the humoral pathway. The second phase (Fig. 40, II) of intensification of the JSPO is evident 120 min (E-G) after intravenous injection of phosphacol (0.3 mg/kg). The amplitude of the ISPO with a frequency of 10 osc/min is sometimes in excess of 1.2 mV. There is also a A-rhythm (4-5 osc/min) with an amplitude of about 2 mV (G). Solitary discharges occur at about this time (D). Ether anesthesia results in complete depression of the infraslow oscillations (H, I); the ISPO is restored after the anesthesia wears off (K). Intensification of the ISPO during the initial phase may be due to the effect of the inhibitor on nonspecific cholinesterase. If inhibited, the

OSCILLATIONS I N AN ISOLATED STRIP O F CORTEX

85 effect on brain function may be more pronounced than if true cholinesterase were inhibited (Desmedt and La Grutta, 1955, 1957). The second phase of ISPO intensification may be caused by the effect of phosphacol mediated through the subcortical centers. As a result of excitation of the latter, the action of phosphacol changes the hormonal level, thus influencing the metabolism of the cortical strip. An injection of phosphacol directly into the periventricular region of the hypothalamus intensifies the ISPO in intact cortex within 20-40 and 150-200 min of the injection. The ISPO's s e e m to be electrical phenomena of a different order with a different electrochemical nature. The more rapid activity, which r e flects the synaptic mechanism of excitation, depends on the presence of afferent impulses. Afferent impulses of cortical neurons do not seem to be required for the generation of infraslow oscillations. Infraslow rhythms appear 1-2 h after isolation, when the more rapid"spontaneous" activity is no longer present. Thus, the origin of infraslow rhythms is not directly related to summation of the postsynaptic potentials. The ISPO tends to reflect unusually prolonged changes in excitability caused by metabolic.shifts in different elements of the cell structure due to changes in the r a t e of the chemical reactions (Zhukov, 1948). These processes a r e maintained humorally. The' chemical factors , in turn, a r i s e after neuroendocrine reaction of the organism to changes in the internal and external environments.

Effect of hypothalamic stimulation on the lnfraslow potential oscillations in isolated strips of cerebral cortex

ISPO in the cerebral cortex may be intensified by stimulation of the hypothalamic area. The hypothalamus is linked to the cortex both by nerve and by extraneuronal connections (Chapt,er V). The "isolated cortex" method may be helpful in elucidatiilg the role of extraneuronal influence of the hypothalamus on ISPO in the cortex. The hypothalamus was stimulated through bipolar electrodes generally implanted in the preoptic or supraoptic nuclei on both sides. The distance between the stimulating electrodes was quite small (about 100 p) in order to achieve maximum local stimulation. In addition, the duration of the stimulus was quite short, 10 msec, thus ensuring its attenuation after penetrating a short distance into nerve tissue (the stimulus did not exceed 3 V). Stimulation of the hypothalamic region resulted in the appearance or intensification of the ISPO in isolated s t r i p s of cortex. Simultaneous recording of the ISPO in both isolated and intact cortex of the opposite hemisphere showed that changes appeared in the strip sooner and were more pronounced. Intensification of the ISPO in the strip became p e r ceptible 20-200 min after the hypothalamus was stimulated. It occurred in 9 cases 17-50 min later, in 5 cases 60-200 min later, and in 3 cases 120-230 min later. Spontaneous electrical activity with a dominant rhythm of 2-3 c/s occasionally appeared in the s t r i p at the same time. The ISPO's were also intensified in the symmetrical intact cortex and

86

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

Fig. 41. Intensification of ISPO in isolated cortex after stimulation of the perichiasmal region. (1)= 60 min after isolation; (2) = l rnin after stimulation; (3) = 32 min; (4)= 65 min; (5) = 105 min; (6) = 150 min after stimulation.

a rhythm of about 3 c/s appeared in the electrocorticogram. The ISPO in the strip generally had a frequency of 3-5 osc/min; a frequency of 12 osc/min was noted in only 2 cases. Fig. 41 shows the intensification of E P O in isolated cortex (A) some time after stimulation of the perichiasmal region. The effect in the strip appeared 32 min (first stage) and 105 min (second stage) after stimulation. Stimulation of the supraoptic region of the hypothalamus likewise intensified the ISPO in intact cortex in two stages: 30-50 min and 80120 min after stimulation; the electrocorticograms also changed at these times. In the experiment shown in Fig. 42, intensification of the ISPO in the strip occurred 24-100 min and 120-200 min after repeated stimulations, 120-215 min later in intact cortex. ECoG's of intact cortex 5 rnin after stimulation revealed a depression of electrical activity. High-amplitude discharges appeared periodically 40-60 min later. After 107 min, low-amplitude rapid activity was dominant and a rhythm of about 1 c/s appeared. During the second-stage (140-225 min later) highamplitude discharges reappeared and the activity became spindle-shaped. We pointed out in Chapters II and III that the effect of intensified infraslow brain activity is manifested in several phases: the first stage, which develops 30-110 min after excitation, and the second stage, which becomes manifest after 130 min. The electrocorticograms likewise change at these times; the high frequency components become intensified in the first stage. However, these stages are not always sufficiently distinct in the ECoG to be identified. "he spindle-shaped type of activity appears in the second stage. These two stages are also reflected in changes in electrical activity in isolated strips of cortex. Moreover, the first stage is more distinct in isolated cortex than in intact cortex. Stimulation of the supraoptic region of 9 e hypothalamus, which also causes neurosecretion, evokes ISPO both in intact and in isolated cortex. This shows that changes in the content of chemically active

OSCILLATIONS IN AN ISOLATED STRIP O F CORTEX

Fig. 42. ISPO in isolated (I) and intact (11) cortex and ECoG of intact cortex (111) after stimulation of the supraoptic nucleus. (I, AX) 32 min after isolation; (B1) 24 rnin after stimulation of the supraoptic nucleus: (Cland D1) 115 and 205 min after stimulation; (11, A2) before stimulation; (Bz, C2, and Da) 9, 127, and 215 min after stimulation; (111, &) before Stimulation; (B3,C3, D3, E3, Fa) 5 , 40, 66, 107, and 225 rnin after stimulation.

87

88

EXCITABILITY O F NEURONS IN THE CEREBRAL CORTEX

substances in the internal environment have a generalized effect on the processes underlying the phenomenon of the infraslow rhythm. These processes become manifest after 120 min or more, in intact cortex and somewhat earlier in isolated cortex due, perhaps, to its greater sensitivity. Electrical activity in intact cortex changes within 2 min of the termination of stimulation. It is probably the result of transmission of impulses along the ascending nerve payhways. After 100 min, when a complex of hormonal factors s e e m s to be included in the reaction, large slow waves appear in the cortex, reflecting the generalized effect of hormonal changes in the internal environment. The existence of a relationship between intensification of ISPO in isolated cortex and hormonal changes is likewise confirmed by experiments that involved the injection with 2 mg/kg of desoxycorticosterone. The irregular rhythm of the ISPO observed in the s t r i p prior to injection of the hormone became accelerated 45 min later, reaching a maximum in 80-90 min. The effect disappeared after 125 min, but then often reappeared after 150-200 min. Sometimes the intensification of the ISPO became apparent within 10 min of the injection. However, injection of hormones is certainly not a "pure" experiment demonstrating the role of this particular factor because it undoubtedly affects other tissues while eliciting a reflex response by the hypothalamus and changing the activity of the other endocrine glands (Ilyina and Tonkikh, 1957). Thus, the parameters of the ISPO in isolated cortex a r e subjected to the influence of the hypothalamus exerted extraneuronally. The mechanism of this influence is probably activated by a shift in the hormonal level. The hormones a r e specific chemical regulators affecting the interaction of different links in the processes of cell metabolism. Their influence is manifested in various enzymes in cellular and tissue structures (Goldstein, 1959; Utevskii, 1959). Experiments on isolated cortex show that the ISPO arise in cortical elements and that they can be altered chemically. Experiments with topical application of serotonin to isolated strips of cortex provide important evidence that the ISPO are intensified by the direct effect of the hormone on cortical structures. The effect was manifested within a few seconds of application of a 0.01% concentration of serotonin. The infraslow oscillations had a frequency of 8-10 per min and lasted about 20 min. However, the possibility of synaptic action of serotonin cannot be ruled out (Brodie and Shore, 1957). CONC IJSIONS

The excitability of cortical neurons undergoes cyclic changes of varying duration equal.to the period of the ISPO. The pattern of the more rapid electrical activity in a circumscribed portion of cortex changes in time with the infraslow rhythm. Potential oscillations in the electrocorticogram riay have a larger amplitude and lower frequency at a maximum of the slow wave than at a minimum. The change in amplitude of the rapid activity at a maximum of the infraslow wave as compared with a minimum does not result from a change in the d.c. level because

CONCLUSIONS

89

this level is much greater than the amplitude of the ISPO. This means that the processes reflected in the infraslow rhythm are of specific significance in the excitability of certain cortical neurons. The relationship between cortical excitability and phase of the infraslow wave is also apparent from the reactivity of the cortex to rhythmic stimulation. Prolonged rhythmic stimulation of a receptor causes periodic impairment of the assimilation of this rhythm by the cortex. The periods of regular and impaired rhythm coincide with the periods of infraslow potential oscillations. The recruiting potentials recorded f r o m the cortical surface after rhythmic stimulation of the nonspecific thalamic nuclei intensify and weaken with a period of 8-10 sec, i.e., with the period of the ISPO. However, the reaction of different elements of the cortex is not uniformly correlated with the phase of the infraslow wave, as shown by analysis of the shape of electrical response by dir e c t stimulation of the cortical surface. Changes in negative potential and especially in the positive deflection that follows it (resulting from excitation of the neuron bodies in layers V and VI) reveal the closest relationship to the ISPO phase. The initial positive potential of the r e sponse does not change at a maximum or minimum of the infraslow wave. This s e e m s to indicate that the activity of the neurons in layers III and IV has no direct connection with the ISPO. In the shape of the recruiting potential and the potential evoked by receptor stimulation, it is chiefly the negative component (dendrite potential) and subsequent positive deflection that change. Not all cortical neurons are participating in the rhythmic ISPO. Hence, the excitability of some nerve elements in the cortex changes in time with the infraslow rhythms, while the excitability of other elements is not related to these slow processes. It is now a fairly well established fact that the ISPO a r e not the result of summation of the more rapid activity; they have another, electrochemical mechanism of origin. They reflect processes that are responsible for periodic changes in neuron excitability and are therefore dependent on the nature of the more rapid manifestations of electrical activity in the cortical neurons. The independent existence of infraslow activity can be demonstrated on an isolated strip of cortex. The electrical activity reflecting the synaptic mechanism of transmission of excitation disappears in deafferented, resting cortex because impulses do not a r r i v e via the afferent pathways. However, ISPO are present and their parameters can be changed by external agents. Infraslow rhythms can be changed in isolated cortex by chemical fact o r s reaching it through the internal environment or by direct application. A cholinesterase inhibitor is one of the effective means of intensifying the ISPO;the effect is manifested both within a few minutes of intravenous injection and some 60-200 min later. The first effect is regarded as the result of the direct action of the inhibitor on tissue cholinesterase; the second is regarded as the result of change in the hormonal interrelations in the internal environment. Stimulation of subcortical structures (preoptic nuclei of the hypothalamus) definitely changes electrical activity in isolated strips of cortex, even though the nerve connections between them are cut. Infraslow rhythms in isolated

90

EXCITABILITY O F NEURONS I N THE CEREBRAL CORTEX

cortex generally have a frequency of 3-5 and sometimes about 12 osc/ min. They change in response to stimulation of the hypothalamus, to injection of some hormonal substances into the blood, and to direct application of the substances to the strip. This signifies that hormonal shifts in the organism play some p a r t in the ISPO phenomenon, thereby agreeing with the data presented in the preceding chapter on the intensification of ISPO in the cortex of animals and man following the response of the central nervous system to "stress". It has already been noted that the intensification of infraslow activity in isolated cortex caused by chemical influences or stimulation of the hypothalamic structures may become evident many minutes or even hours later. This "late" intensification is frequently oscillatory in nature, appearing in two stages: after 30-80 min and after 100-200 min. At this time an ECoG can be recorded from isolated cortex with a characteristic oscillation frequency of about 3 c/s. These and other data were the basis of the theory that the brain has a so-called slow control system (cf. Chapter VII). The interaction between elements of this slow control system of the brain and the rapidly acting system at the cortical level is brought about through the influence of those cortical elements (characterized by ISPO)on the excitability of other elements, whose function is related solely to the rapid forms of activity.

91 CHAPTER V

THE ROLE O F CERTAIN SUBCORTICAL STRUCTURES IN THE APPEARANCE O F INFRASLOW POTENTIAL OSCILLATIONS IN THE BRAIN INFRASLOW POTENTIAL CHANGES IN NUCLEI O F SUBCORTICAL AND UPPER BRAIN STEM STRUCTURES

Analysis of the pathways by which the ISPO a r e intensified and the lag in appearance of the response to stimulation of the hypothalamus show that neurohumoral relations play a role in the phenomenon of infraslow oscillations. The hypothalamus is a structure that controls the level of activity of the endocrine glands. It performs this function in accordance with the information reaching it both from the cortex and from the subcortical structures (reticular formation of the brain stem, amygdaloid nucleus, thalamic structures) and with the direct influences exerted by internal environmental factors. Many forms of hypothalamic regulatory influences are realized through neural and neuro-endocrinal pathways. These influences a r e exerted both through structures of the autonomic nervous system (Orbeli, 1935) and neurohormonal connections with the adenohypophysis. These functions of the hypothalamus affect to some extent the "metabolic level" of central nervous tissue and the local chemical gradients therein. The result is probably reflected in rhythmic infraslow potential oscillations.

Connections and function of the hypothalamus Owing to its numerous direct and indirect connections with the cortex and with many other divisions of the central nervous system, the hypothalamus participates in a variety of manifestations of nervous activity, particularly those related to autonomic functions. ,The hypothalamic structures presumably play an important p a r t in the integrative activity of the brain. A great many comprehensive surveys have been published in recent years on the connections, structure, and function of the hypothalamus. Therefore, this chapter will discuss only briefly some of the most important views, particularly those bearing most closely on the ISPO phenomenon. There are two-way connections between the hypothalamus and cortex, basal ganglia, rhinencephalon, thalamus and reticular formation. The cerebral cortex is connected with the hypothalamus through the hippocampus, thalamus, amygdaloid nuclei, globus pallidus, and probably some other pathways. There is some relationship between several regions of the cortex and individual regions of the hypothalamus. The connections between the hypothalamus and hippocampus are essential for execution of the integrative activity of the brain associated with the nervous mechanism of emotions. The connections included in the

92

ROLE OF CERTAIN SUBCORTICAL STRUCTURES

so-called "hippocampal circuit" (Papez, 1937) play an unusually imPortant role here. The "hippocampal circuit" is formed by gyrus c i n s l i , hiPPocamPus, fornix cerebri, hypothalamus, and anterior complex of nuclei Of the thalamus and gyrus cmguli. The abundance of connections between the various structures of the "hippocampal circuit" and other divisions of the brain (Nauta, 1956) creates unusual opportunities for the organization of self-regulatory processes in the brain, e g . , the connections of the hippocampus with the reticular formation and with many thalamic and hypothalamic nuclei not included within Papez' circle. Other examples a r e the connections between the structures of the rhhencephalon and hypothalamus effected through projections of the amygdaloid nucleus, the role of the periventricular system of fibers in connecting the hypothalamus with the thalamic nuclei and the central gray substance around the aqueduct of Sylvius. The connections between the central gray substance and hypothalamus a r e also realized through the radial fibers of the tegmentum and peduncles of the mammillary bodies (Adey et al., 1956). The multiplicity of connections between the hypothalamic structures and other divisions of the brain confirms the importance of the hypothalamus in integrating somatic and visceral activity, which has mostly a homeostatic character. The hypothalamus influences the following systems and functions in the body: (1) cardiovascular, (2) thermoregulatory, (3) digestion, excretion, (4) water metabolism, (5) sleep and wakefulness, (6) respiration, (7) endocrine glands, (8) hematopoiesis, (9) regulation of metabolic processes, and others. The organization of adaptation (primarily autonomic and humoral) reactions is largely effected*&rough the hypothalamus. The commonest methods of studying hypothalamic functions include: (1)destruction of the structure of individual areas (e.g., by electrocoagulation), (2) electrical stimulation of different nuclei, (3) injection of various substances into different parts of the hypothalamus or third ventricle, (4) recording of electrical activity and other manifestations of activity of the hypothalamic nuclei in response to peripheral stimulation or to stimulation of other central nerve structures. The most local effect on the hypothalamic nuclei is achieved by electrical stimulation carried out with bipolar microelectrodes placed close together (50-150 and by weak stimulation with square pulses of current of short duration. Strong stimulation may involve other neural structures and conducting pathways, just as the injection of chemical substances involves groups of hypothalamic structures. On the other hand, experiments with intravenous injections of substances suggest that it is difficult to draw clear-cut conclusions regarding the direct effect of these substances on hypothalamic functions because different receptors of the vascular bed and other divisions of the brain a r e involved in the reaction, and these, in turn, influence the hypothalamus. The method of stimulating different nerve structures and recording hypothalamic activity is often combined with the method of in vivo transection of the brain at different levels. The following a r e indicators of change in hypothalamic activity:

INFRASLOW P O T E N T I A L CHANGES I N N U C L E I

93

(1) general sympathetic and parasympathetic effects, (2) shifts in the course of autonomic reflexes, (3) dynamics of change in the hormone content of the blood, (4) change in excitability of other divisions of the central nervous system, (5) electrical activity of the cerebral cortex and other divisions of the central nervous system, (6) electrical activity of the hypothalamus, (7) changes in higher nervous activity (judged by conditioned reflexes). A detailed account of the r e s e a r c h done on physiology of the hypothalamus can be found in the articles of Ingram (1960), Harris (1960), and Aleshin (1960). We shall therefore present h e r e only a small portion of the data relating to the role of the hypothalamus in neurohormonal relations. The central nervous system regulates the production of hormones ultimately by means of the hypothalamus and its influence on the pituitary. The production and secretion of hypophyseal hormones is mostly centered in certain regions of the hypothalamus. The production of adenohypophyseal hormones is controlled chiefly by the central hypothalamus, while the adrenocorticotropic function is regulated by the posterior hypothalamus. The anterior hypothalamus is connected with the posterior lobe of the pituitary. Two regulatory pathways are thought to be possible: nervous and hormonal. The most popular view is that neurosecretory material is released in the hypothalamus and then transferred to the pituitary (along the nerve fibers o r through the portal blood vessels), determining the level of secretion of the pituitary hormones proper (Harris, 1955, 1960). The r o l e of the hypothalamus in the mechanisms of nonspecific adaptation to changing conditions in the internal and external environments of the body consists in mobilizing both the somatic and the autonomichormonal defense forces. Electrical activity of the hypothalamus is intensified during the development of a "stress" reaction in the organism (Porter, 1952). Transection of the brain stem at the bulbar level and at the pontine level does not prevent the hypothalamus from becoming involved in the "stress" reaction; only higher transection is effective in this respect. Injury to the posterior hypothalamus prevents the appearance of ''stress" reactions, but injury to the anterior hypothalamus does not have this effect. It is interesting to note that the "stress" state is characterized by production of the antidiuretic hormone, which is not prevented by hypophysectomy, a fact which suggests that this hormone originates in the hypothalamus (cf. below on the proc e s s e s of neurosecretion in the hypothalamus). The activity of the hypothalamus is an indispensable link in several behavioral reactions. It is particularly marked in sexual behavior, sensations of hunger and thirst, emotional reactions (anger, fear), defensive behavior (flight), alternation of sleep and wakefulness. Influences brought to bear on the hypothalamus cause changes in higher nervous activity (Deryabin, 1946). Stimulation of the hypothalamus may inhibit o r intensify the food and defense conditioned reflexes; the region that inhibits the food conditioned reflex facilitates the defense reflex (Lissfi, 1955). Facilitation of the conditioned defense reflex is manifested after stimulation of the posterior hypothalamus (Livanov,

94

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

1959). Intensification of electrical activity can be observed in the hypothalamic structures themselves (in the lateral and medial regions) after formation of the food reflex (Lyubimov, 1958). One of the principal functions of the hypothalamus seems to be autonomic and metabolic performance of acts of higher nervous activity (Anokhin, 1958a and b). In addition, the hypothalamus plays a leading role in the creation of emotions which, in turn, are involved in the mechanisms of fixation of temporary connections. The "emotive background", according to Pavlov, is important in the establishment of conditioned reflexes. "The main impulse to cortical activity comes from the subcortex. If the emotions are excluded, the cortex is deprived of its main source of strength" (Pavlov,Complete Works, Vol. 1, p. 268).

Neurosecretory mechanisms of the hypothalamus The supraoptic and paraventricular areas of the hypothalamus contain a mass of nerve cells capable of elaborating neurosecretion. The size p f the nucleus and cytoplasm of the neuron body changes reversibly in the process of neurosecretion. At the same time there is a redistribution in the neuron of neurosecretory material, which may be released into the blood or be transported along the axon. The neurosecretory material is found in the cell near the nucleus, but is closer to the periphery in the axon. The granular intracellular secretory material of the hypothalamus is transported to the neurohypophysis where it is visible in the form of masses of granules. Transection of the tract between the supraoptic area and pituitary results in an accumulation of neurosecretion in the hypothalamus. The amount of neurosecretion and elaboration of hormones change in relation to each other. The secretory activity of the hypothalamic neurons is the principal mechanism by which the hypothalamic nuclei regulate the pituitary. The steadily increasing volume of research indicates that the neurosecretory fibers of the hypothalamus need not proceed only to the pituitary, that they can terminate in different regions of the brain. These endings a r e found, for example, in the a r e a of the septum pellucidum, mesencephalon, and w a l l s of the lateral ventricles. Neurosecretion can be released into the ventricles up to the lateral ventricles by transfer along the long fibers from the paraventricular nuclei (Wilson et al., 1957). There are peculiar bipolar neurons (Stutinsky, 1953) which secrete drops of a colloid substance into the ventricles. One of the poles of such bipolar neurons is found in ependymal cells and it penetrates through the wall of the third ventricle. Ford and Kantonnis (1957) found neurosecretory cells with axons in many p a r t s of the hypothalamus passing along the walls of the third ventricle. Barry (1954a, b) observed the transport of secretion to the lateral ventricle and found (in mouse brain sections) fibers proceeding from the hypothalamus to the walls of the lateral ventricles. These fibers, in his opinion, are neurosecretory pathways. Secretory structures have been observed at the level of the posterior commissure and underneath it. There are also indications that the neuroglia of the hypothalamus has a secretory

INFRASLOW POTENTIAL CHANGES IN NUCLEI

95

function (Collin, 1956). For example, glial secretory cells have been found in the mammillary region. Thus, there would seem to be two kinds of hypothalamic neurosecreT tion: secretion of granular material proceeding t o the pituitary and secretion of colloid material ascending along the ventricular walls. The nature of the secretory material has not been accurately established. It has been found to differ from vasopressin, histamine, acetylcholine, epinephrine, norepinephrine, and serotonin. Gillemin et a l . (1956) discovered that it has sulfhydryl groups in a form bound with protein. It is said to resemble the antidiuretic hormone (McCann, 1957). B. and E. Scharrer (1944) believe that the pituitary hormones can travel toward the hypothalamus. The hypothalamus occupies a special place among the various brain structures in its content of active substances (histamine, norepinephrine, 5 -hydroxytryptamine, acetylcholine, and substance P). They a r e not distributed uniformly in the different parts of the brain. Norepinephrine, 5-hydroxytryptamine, and substance P are also found in other parts of the brain, especially in the nonspecific structures, e.g. they are more abundant in the medial nuclei of the thalamus than in the lateral nuclei, and they are less abundant in the reticular formation than in the hypothalamus. The relative distribution of active substances in the brain is shown in Table II. T A B L E I1 RELATIVE DISTRIBUTION OF ACTIVE SUBSTANCES IN THE BRAIN

Parts Of brain

White substance of cerebral hemispheres Gray substance of cerebral hemispheres Motor cortex Visual cortex Olfactory cortex Corpus callosum Caudate nucleus Olfactory bulbs Hippocampus Medial thalamus Lateral thalamus Anterior hypothalamus Posterior hypothalamus Central gray substance Cerebellum F oor of fourth ventricle Area postrema

P, units/g

5-hydroxytryptamine, m&kg

Norepinephrine, mw/&

Cholinacety.?hse, % of amount in thalamus

3.6

0

50

-

4.8

29

-

71

19 7.3 29 5.9 46 5.5 15 11 8.4

21

180 40 120 80 60 50 40 240 1030

71 58 81 26 127 55 108

0

16 0 0

48 45 67 0

70

220

22

255

1030

76

280

1030

0

70

1.6 45 290

1030

98 215

100

1040

15

96

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

The Table was compiled from the data of Amin et a l . (1954), Vogt (1954), and Feldberg and Vogt (1948).

The greatest concentration of 5 -hydroxytryptamine is found in the hypothalamus and in the a r e a postrema. The latter contains chemoreceptors that are very rich in glia but deficient in nerve cells. The larger content of active substances in this area is apparently due to neuroglial function. And it is also possible that the neuroglia may play a similar role in other parts of the brain as well. The high content of biological active substances in the hypothalamus suggests that it contains chemically sensitive elements capable of responding to changes in the internal environment of the organism and translating them into adequate forms of behavior through appropriate nervous activity.

INFRASLOW POTENTIAL CHANGES IN NUCLEI

97

Topopaphy of infraslow potential oscillations in nuclei of the thalamus, hypothalamus and brain stem Im P lantation of electrodes and his to logica 1 contro 1. Metallic (rigid, chemically inactive metal, e.g., nichrome or tungsten) electrodes about 10 p in diameter may be used to derive potentials from subcortical nuclei. In the case of blpolar derivation, two such electrodes a r e fastened together with insulating lacquer in such a way that the tips a r e 30-100 p apart. Thus both points may record from the same nucleus. The electrodes a r e implanted through trepanations. Ten days after the operation, the injury to the structures caused by insertion of the electrodes will no longer effect the results of the experiment. Histological Fig. 43. Map of distribution of ISPO in different subcortical structures of the rabbit. AC = anterior commissure; AD = anterodorsal nucleus of the thalamus; AHA = anterior hypothalamic area; AM = anteromedial nucleus of the thalamus; AMYG = amygdaloid nucleus; APV = anterior paraventricular nucleus of the hypothalamus; AQ = aqueduct of Sylvius; ARC = arcuate nucleus of the hypothalamus; AV = anteroventral nucleus of the thalamus; BC = brachium conjunctivum; BIC = brachium of the inferior colliculus; BSC = brachium of the superior colliculus; C = caudate nucleus; CC = corpus callosum; CG = central gray substance; CM = centromedian nucleus; CORT = cortex; CS = corticospinal tract; CSC = commisure of the superior quadrigeminal body; CSG = nucleus of Darkshevich; DBC = decussation of the brachium conjunctivum; DHA = dorsal hypothalamic area; DM = dorsomedid nucleus of the thalamus; DMH = dorsomedial nucleus of the hypothalamus; DTD = dorsal tegmental decussation; EC = external capsule; FC = foramen caecum; FF = fimbria fornicis; FM = foramen of Monro; FX = fornix; G = mammillotegmental tract; GP = globus pallidus; HM = medial nucleus of the habenula; HPC = hippocampus; HT = habenulopeduncular tract; IC = internal capsule; IMD = intermediodorsal nucleus; INF = infundibulum; IPN = interpeduncular nucleus; LA = anterolateral nucleus of the thalamus; LGD = dorsal part of the lateral geniculate body; LGV = ventral part of the lateral geniculate body; LHA = lateral hypothalamic area; LL = lateral loop; LLN = lateral loop nucleus; LM = medial loop; LP = posterior lateral nucleus of the thalamus; LPO = lateral preoptic area; LT = lamina terminalis; M = midline nuclei; ME = median eminence; MG = medial geniculate body; ML = lateral mammillary nucleus; MLF = medial longitudinal fasciculus; MM = medial mammillary nucleus; MPO = medial preoptic area; MT = mammillothalami-ctract; NC = central nucleus of the thalamus; NRF = nucleus of reticular substance; OCH = optic chiasma; OT = optic tract; PC = posterior commissure; PED = base of cerebral peduncles; P F = area near fornix; PHA = posterior hypothalamic area; P M = peduncles of mammillary body; PMA = premammillary area; P N = nucleus pontis; PONS = brachium pontis; PPO = paraventricular preoptic nucleus; PTH = prethalamic nucleus; PTN = peristriatal nucleus; PUL = pulvinar; PUT = putamen; P V = paraventricular nucleus of the hypothalamus; RET = reticular nucleus of the thalamus; RF = reticular substance; RN = red nucleus; SC = superior colliculi; SCH = supraoptic nucleus; SG = suprageniculate body; SM = stria medullaris; SMA = supramammillary area; SN = substantia nigra; SO = supraoptic nucleus; SOD = diffuse supraoptic nucleus; SP = septum pellucidum, STH = subthalamic area; V = ventral nucleus of the thalamus; VA = anterior ventral nucleus of the thalamus; VEN = ventricle; VL = lateral ventral nucleus of the thalamus; VM = ventromedial nucleus of the thalamus; VMH = ventromedial nucleus of the hypothalamus; VML = ventral medullary lamina; VPL = posterolateral ventral nucleus of the thalamus; VPM = posteromedial ventral nucleus of the thalamus; VTD = ventral tegmental decussation; 111 = oculomotor nerve; IIIN = oculomotor nucleus: IIrV = third ventricle.

98

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

control shows that there is a thickening of glial tissue in a radius of about 100 around me electrodes. The effect of minor injury in the adjacent (noninvestigated) areas is slight enough to be ignored. We used a stereotaxic instrument designed by Mescherskii (1959), who was kind enough to send u s his sketches, to implant the electrodes. We used a rabbit brain atlas with a system of coordinates worked out by Sawyer et al. (1954). The position of the electrodes was histologically controlled; using section 10 p thick. Besides the subcortical electrodes, silver disc electrodes 4 mm in diameter with centers 6 mm apart were implanted in each rabbit on the surface of the sensorimotor region of one or both hemispheres. The potentials were recorded with two amplifiers: d.c. to derive the ISPO and a.c. to derive the ECoG. Electrical stimulation was effected with square pulses, each 10 p s e c long, with a frequency of 50 c / s and amplitude of 2-4 V, which did not elicit any motor reactions. In certain experiments, to be discussed below, stimulation w a s at a frequency of 6 and 300 c / s .

Topogvaphy of infraslow potential oscillations (ISPO) in the diencephalon and brain stem The experiments with ISPO in the nuclei of the thalamus and hypothalamus were performed jointly with Kol' tsova. Fig. 43 contains schematic representation of sections of the brain in which the tips of the electrodes were implanted. The signs (+) and (-) indicate the presence or absence, respectively, of ISPO in these nuclei (the topography'of the recordings within the nuclei was not taken into consideration, and the number of signs signifies only the number of experiments with recordings from a given nucleus). The map was compiled from experiments on 40 rabbits and 82 eiectrode implantations in different nuclei. Infraslow rhythms were absent in the thalamic nuclei of 32 wakeful rabbits subjected to no influences other than the operation of implanting the electrodes (Aladjalova and Koltsova, 1958). Nor were they found in the central gray substance, brain stem reticular formation, or h ippocampus . Fig. 44 presents recordings from both the lateral and the medial nuclei of the thalamus and from the brain stem. The nature of the electrical activity recorded from the same electrodes indicates that the nuclei preserved their function. ISPO's were not detected in the thalamic nuclei either by bipolar or by unipolar recording. On the other hand, ISPO's were recorded quite frequently in the hypothalamus of intact animals, but they were irregular. Different areas of the hypothalamus were investigzted in 22 rabbits. ISPO's were recorded in 14 but were absent in 8 animals. They were most frequently found in the anterior and medial hypothalamus. In the same group of rabbits, ISPO's were recorded in the sensorimotor cortex in 28 out of 40 cases (several pairs of electrodes were placed on the cerebral,cortex). The ISPO in the hypothalamus of intact animals had a frequency of

FACTORS INCREASING INFRASLOW POTENTIAL OSCILLATIONS

99

Fig. 44. Absence of ISPO in the nuclei of the thalamus and reticular formation in intact animals and the presence of ISPO in the nuclei of the hypothalamus and in the cortex. Left: recording from nuclei of the thalamus and reticular formation. Above: ISPO (absent), below: ECoG. Right: recording of ISPO from cerebral cortex and nuclei of the hypothalamus (symbols of the nerve structures a r e the same as in Fig. 43).

I

6-10 osc/min and a n amplitude of 0.2-0.5 mV. This rhythm has the same sequence as the A-rhythm in the sensorimotor cortex. The latter in these experiments was more pronounced on the side where the electrodes were implanted in the hypothalamus, a possible result of stimulation by the electrodes. Ether anesthesia causer2 total disappearance of the ISPO both in the hypothalamus and in the cortex.

FACTORS INCREASING INFRASLOW POTENTIAL OSCILLATIONS IN SUBCORTICAL STRUCTURES

Electrical stimulation of the hypothalamus and thalamus

Various areas of the hypothalamus (21 experiments) were stimulated several times for 30 s e c to 3 min. From 2 to 20 min after stimulation, the ISPO became intensified in the nuclei stimulated. They also became intensified in the sensorimotor cortex 2-20 min after stimulation. Repeated stimulation resulted in synchronization of the ISPO in symmetrical areas of the cerebral cortex and in the hypothalamus. Fig. 45 presents two-channel recordings of ISPO's. The electrograms show that the potential oscillations were not synchronous in the various areas. Stimulation of the dorsomedial nucleus of the hypothalamus three times (about 30 sec each) evoked synchronous infraslow oscillations (after 120 min) in all the areas with a frequency of 5 per min and an amplitude ranging from 1-1.5 mV. Synchronization of oscillations in the ECoG developed simultaneously with synchronization of the JSPO. Fig. 17 ( c f . Chapter III) presents ECoG' s made at different intervals of time after repeated stimulation of the ventromedial nucleus of the hypothalamus. High-amplitude synchronous waves (3, 4) appeared after such stimulation and changed in

100

ROLE OF CERTAIN SUBCORTICAL STRUCTURES

Fig. 45. Intensification and synchronization of ISPO in the sensorimotor cortex and in the hypothalamus after stimulation of the dorsomedial nucleus of the hypothalamus. (I) before stimulation; (11) after stimulation. A = simultaneous recording from stimulation of the dorsomedial nucleus of the hypothalamus (above) and f r o m the left sensorimotor cortex (below); B = recording from symmetrical ,areas of the cortex of both hemispheres (above = right hemisphere, below = left),

time with the infraslow waves. The cortical ISPO became intensified after stimulation of the hypothalamus, but they diminished in frequency from 8-5 osc/min. ISPO intensification after hypothalamic stimulation i s very persistent; it may remain for two or more hours afterwards. Electrical activity of the nucleus of the stimulated hypothalamus inc r e a s e s periodically (cf. Chapter VII). High-amplitude activity appears in the record of the nucleus (Fig. 46) while the infraslow waves may

Fig. 46. Recording of the dorsomedial nucleus of the hypothalamus (I) and infraslow oscillations therein (II) after stimulation of the nucleus. (1)before stimulation; (2) after 30 sec of stimulation; (3) after additional stimulation.

become big. The effect of ISPO intensification was observed even after stimulation of the medial preoptic a r e a in which there were no ISPO's prior to the experiment. Three minutes of stimulation intensified the ISPO both in the area stimulated and in the cortex. The TSPO's occasionally remained unaltered after a single stimulation; they became intensified only after long and repeated stimulation and persisted for several hours. The changes in parameters of the ISPO in the hypothalamic nuclei

FACTORS INCREASING INFRASLOW POTENTIAL OSCILLATIONS

101

had an interesting feature,-i.e., after the nuclei were stimulated, ISPO's of an appreciably higher frequency than in the cortex often appeared in these nuclei. F o r example, stimulation of the dorsomedial nucleus (twice, 50 sec each time) resulted in the recording of a rhythm of about 18 osc/ min with an amplitude of 0.1 mV. The ISPO in the hypothalamus then virtually disappeared, but again became intensified 2 h after stimulation with a frequency of about 24 osc/min and an amplitude of 0.2 mV. There w a s an ordinary frequency of about 6 osc/min in the sensorimotor cortex 15 min after stimulation, but it likewise increased (to 1 2 osc/min) after 2 h, although it did not coincide with the frequency of oscillations in the dorsomedial nucleus of the hypothalamus. The frequency of the ISPO increased from 6-8 to 10-12 and sometimes 14-20 osc/min in the hypothalamic structures of the premammillary and medial preoptic areas, dorsomedial and ventromedial nuclei after they were stimulated. This increase in ISPO frequency is highly characteristic of the hypothalamic region and is probably related to the peculiarities of the metabolic processes in the hypothalamus. We sometimes observed reciprocal relations between the frequency of the ISPO in the hypothalamus and in the ipsilateral cortex. For example, after stimulation of the medial preoptic area, the frequency increased f r o m 6 to 1 2 osc/min, but decreased in the ipsilateral cortex from 9 to 5 osc/min. Consequently, the frequency of the ISPO may increase in the thalamus while decreasing in the cortex. However, the oscillations become less regular and a picture of synchronous activity appears in the ECoG. The difference between the change in ISPO frequency in the cortex and that in the hypothalamus shows that under these conditions ISPO's a r i s e in the cortex not as a result of the direct transmission of potential rhythm from the hypothalamus, but through some intermediate link. This link appears to be a chemical factor of unknown nature which enters the blood upon stimulation of the hypothalamic nucleus and influences those processes in the cortex which a r e reflected in the cortical ISPO. The temporary correlations between the observed phenomena constitute evidence in favor of this assumption. The effect appears not immediately after stimulation, but after a substantial interval of time. The effect may be noted at two different periods, the immediate one within 2-15 min of Stimulation, the remote one after an interval of about 40-100 and 120-240 min. ISPO's are not distinct in the thalamic nuclei of an intact animal (Fig. 44). Stimulation of the hypothalamus too generally failed to evoke ISPO in the thalamic nuclei. Only once did weak ISPO's appear in the dorsomedial nucleus of the thalamus after stimulation of the dorsomedial nucleus of the hypothalamus, but they could not be detected in the lateral ventral nucleus of the thalamus of the same rabbit. Direct stimulation of the thalamic nuclei also failed in most cases to evoke ISPO's therein. A total of 17 experiments were performed with stimulation of different nuclei of the thalamus. Electrical stimulation of the thalamic nuclei depressed the =PO in the cortex. In Fig. 47, the + sign designates subcortical nuclei whose stimulation resulted in intensification of the ISPO therein. The - sign designates

102

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

Fig. 47. Changes in ISPO in different subcortical nuclei after electrical stimulation thereof. + = appearance or intensification of ISPO after stimulation; - = absence of effect; 0 = complete depression of ISPO; @ = partial depression .(cf. Fig. 43 for the names of the ne me structures).

the nuclei whose stimulation failed to produce any changes therein. The circles designate the effects of depression of the ISPO after stimulation (the number of signs within each nucleus indicates the amount of recordings from it without taking into account the intranuclear location).

Stimulation of the taste receptor Stimulation of the taste receptor by applying a bitter substance to the tongue intensified the ISPO both in the cortex and in the hypothalamic nuclei, the effect being manifested f i r s t (within 1-10 min) in the hypothalamic area, then (after 20-60 min) in the cortex. It was most pronounced in the cortex at the 45th min. At the same time synchronous high-amplitude oscillations appeared periodically in the recording of the hypothalamus. Intensified ISPO's were also noted in the thalamic nuclei of three rabbits whose taste receptor had been stimulated. They appeared 8 min after the start of stimulation; intensification was evident at the 20th min in the cortex as well (Fig. 48). Intense stimulation of the taste receptors apparently increases the activity of the hypothalamic area. As a result, a factor (probably chemical) is produced which begins to act within 15-30 min on the metabolic processes in the cortex.

FACTORS INCREASING INFRASLOW POTENTIAL OSCILLATIONS

103

Fig. 48. Intensification of ISPO in different structures after stimulation of the taste receptor. (I) ventromedial nucleus of the hypothalamus; (11) dorsomedial nucleus of the thalamus; (111) sensorimotor cortex of another animal. The figures on the left indicate the time the effect. appeared after stimulation of the receptor.

Effect of strychnine

Strychnine is specific in its effect on the nervous system. It is assumed to affect synaptic transmission in particular. Either application of the substance to the cortex or injection of it into the blood increases the excitability of the nervous system to the point where convulsions set in and causes characteristic potential oscillations to appear in the electrical activity of the cortex. As noted above, the mechanism of action of strychnine consists in blocking the inhibitory synapses. Murphy and Gellhorn (1945), using the method of strychnine neuronography to investigate the connections between the hypothalamus and c o r tex, obtained interesting data that may be helpful in interpreting some of the results of our investigations. These authors found that highamplitude discharges appeared in the dorsomedial nucleus of the thalamus 10 min after strychnine was injected into the posterior hypothalamus, in the sensorimotor cortex of the ipsilateral side after 30 rnin and in the contralateral side 7 min later. Discharges did not appear in the ventrolateral nucleus of the thalamus. Discharges appeared in the ipsilateral cortex 3 min after strychnine w a s injected into the dorsomedial nucleus of the thalamus, in the ipsilateral half of the posterior hypothalamus after 10 min, and in the contralateral half after 1 2 min. Discharges also appeared in the ventrolateral nucleus of the thalamus (after 60 and 120 min). The authors concluded from their experiments that connections exist between the afore-mentioned structures, with excitation possibly being transmitted from the hypothalamus to the dorsomedial nucleus through fibers proceeding along the w a l l s of the third ventricle. However, it subsequently spreads from the dorsomedial nucleus of the thalamus to the cortex through the nonspecific thalamic projections. They also concluded that there is a direct pathway to the hypothalamus from the ventrolateral nucleus of the thalamus which can be regarded as an intermediate link in the process of activating the

104

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

hypothalamus following stimulation of the receptors. There may be some objections to this conclusion on methodological grounds since other ways of realizing the effect of strychnine are possible (injection of strychnine into the posterior hypothalamus may excite the adjacent nuclei and fibers). In addition, the effect appears some time afterwards, suggesting that it is mediated (through the nervous mechanism) rather than that there is a direct impulse connection between the afore-mentioned p a r t s of the brain. We must also take into account Grundfest's statement that all the neuron structures do not react to strychnine due to the nature of their synaptic organization. We must point out in connection with the foregoing data that the intensification of ISPO after the administration of strychnine was not uniform in the different brain structures and that it occurred at the same intervals of time that were noted by Murphy and Gellhorn (1945). ISPO in the cortex following the injection (subcutaneous) of strychnine at first went through a depression phase during which convulsions occurred and then, 20-80 min after the injection, they became intensified. Meanwhile ISPO also appeared in some subcortical structures. The effect was most pronounced in the hypothalamic area, especially in the dorsomedial and premammillary nuclei of the hypothalamus and even in the medial nuclei of the thalamus. Even daily injection of strychnine for three days, after which the ISPO were greatly intensified in the ventromedial nucleus of the hypothalamus and cortex and appeared

Fig. 49. Effect of strychnine (injected subcutaneously) on ISPO in different nuclei of the subcortex. (I) + = intensification of ISPO; - = absence of the effect (names of the nerve structures a r e the same as in Fig. 43); (11) appearance of the effect (from the ISPO) in the ventromedial nucleus of the hypothalamus (A), in the intralaminar nuclei of the thalamus (B), in the lateral thalamus (C), and in the sensorimotor cortex (D). Figures on the left indicate time (in min).

FACTORS INCREASING INFRASLOW POTENTIAL OSCILLATIONS

105

Fig. 50. Increased frequency of ISPO in t h e hypothalamus and cortex after the injection of strychnine. (A) premammillary region of the hypothalamus: a = before injection; b = 17 min after intravenous injection of 0.5 mg/kg of strychnine; c = 25 min l a t e r ; d = 80 min later; (B) cortex in the same experiment: a = before injection; b = 40 rnin after injection; c = 47 min later; d = 60 min later,

in the nonspecific nuclei of the thalamus, failed to elicit ISPO in the ventrolateral thalamic nucleus (Fig. 49, 11). Fig. 50 illustrates the appearance of a regular rhythm of 8-9 osc/min with an amplitude of about 2.5 mV in the hypothalamus (premammillary region) 17 min after intravenous injection of strychnine (0.5 mg/kg). This intensification phase was preceded by a phase of depressed activity in the hypothalamus and development of convulsions. After 40 min the ISPO became intensified in the cortex (B) as well, the frequency of the rhythm in the cortex increasing from about 8 to 16 osc/min. They decreased an hour later both in the hypothalamus and in the cortex. The frequency of the ISPO observed in the hypothalamic nuclei after the injection of strychnine ranged from 5-9 per min; there was an occasional temporary increase to 12-16 per min. Frequencies of 3 - 6 per rnin were noted in the medial nuclei of the thalamus. The effect of ISPO intensification w a s observed sooner in the nuclei of the hypothalamus than in the cortical p a r t s of the brain. It was then manifested i n the medial nuclei of the thalamus and, a little later, in the sensorimotor cortex. Following intravenous injection of strychnine, changes i n the parameters of the ISPO in the hypothalamic a r e a preceded changes in the cortex by 10-30 min. Accordiqg to Murphy and Gellhorn (1945) discharges appear in the cortex at approximately the same interval after

106

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

local injection of strychnine into the hypothalamic area. The intensification of ISPO in the cortex noted in our experiments was presumably a phenomenon mediated through excitation of the hypothalamic area. The relatively long intervals of time required for the ISPO to appear in the cortex as compared with the hypothalamus and the similar length of time required for discharges to appear in the cortex after local injection of strychnine into the hypothalamus (Murphy and Gellhorn) suggest that there may be a special mechanism responsible for intensifying the ISPO without the transmission of excitatory impulses from the hypothalamus. Nor can one rule out the possibility of transmission of influences from the hypothalamus through the neurosecretory apparatus connected with fibers proceeding from the hypothalamus in the walls of the third ventricle. This pathway is close to the midline structures of the thalamus and it may thus exert a more local hormonal influence on the superincumbent nerve structures than through the hypothalamohypophyseal system. Consequently, strychnine does not affect the ISPO in all the nerve structures (Fig. 49). These oscillations are intensified chiefly in the torte.;, nuclei of the thalamus, and medial thalamus. Depression of the cortical ISPO occurs along with convulsions concomitant with an aperiodic shift in d.c. potential in a negative direction. The ISPO becomes intensified several minutes later. Thus, the effect of strychnine, too, judging from the infraslow rhythm, is manifested f i r s t in the hypothalamic structures. Strychnine is not the only pharmacological agent that intensifies the ISPO in the subcortical structures. The effect of cholinergic substances will be examined on p. 111. SOME EXAMPLES O F ISPO IN THE CENTRAL GRAY MATTER AND IN THE RETICULAR FORMATION O F THE BRAIN STEM

Ascending connections of the reticular formation The diffuse reticular system of the brain stem has an activating and inhibitory effect (Magoun, 1958) on the entire surface of the cortex and thus influences the state of central excitation (Livingston, 1957). A number of cortical fields (somatosensory, parieto-occipital cortex, some portions of the temporal lobe, orbital surface of the frontal lobe, gyrus cinguli, and insular cortex) a r e projected to reticular formation structures, where corticofugal and afferent impulses coming through the collaterals from all parts of the body converge and all kinds of corticofugal influences interact. Some of the cortical fields mentioned above increase reticular formation activity, while others inhibit it. The cortical fields projected to the stem may set in motion an alternation of excitatory and inhibitory processes in the reticular formation, varying in duration from tenths of a second to several seconds. Therefore, the cortex is not only subjected to excitatory or inhibitory influences from the reticular formation, but itself contains regulatory mechanisms

SOME EXAMPLES O F INFRASLOW POTENTIAL OSCILLATIONS

107

that may affect the level of activity within the reticular formation and thereby determine the state of central activity. Receiving extensive information from different receptors and regulating pulses f r o m the cortex, the reticular formation exerts a diffuse influence through the ascending and descending pathways on other structures, including the sensory pathway. The books of Brodal (1957), Magoun (1958), and Rossi and Zanchetti (1960), include detailed surveys of the structure and physiology of the reticular formation. The important thing for u s to stress is the presence of ascending pathways from the reticular formation to the nuclei of the thalamus (chiefly the nonspecific ones) and the pathways to the nuclei of the hypothalamus. There is some information in the literature on the humoral mechanism of transmission of ascending reticular formation influences to the cor tex. For example, Ingvar (1955a) has described the influence of the reticular formation on deafferentiated isolated cortex. Stimulation of the reticular formation may be presumed to result in the formation of active humoral factors. In experiments with cross blood circulation, Purpura (1956) showed that stimulation of an animal's reticular formation causes humoral activation of the cortex of the recipient. The reticular formation mediator passes through the blood-brain b a r r i e r of both animals, though mixing with a great volume of blood nevertheless remains effective. These humoral effects a r e possibly mediated through the hypothalamus, which is involved in the reaction to stimulation of the reticular formation. Features of the ISPO in the reticular formation and central gray matter

ISPO's were not found in the reticular formation o r central gray substance around the aqueduct of Sylvius in intact, resting animals. Similarly, ISPO's generally failed to appear in the reticular formation after single stimulation of the reticular formation either with a lowfrequency (50 c/s) or high-frequency (300 c/s) current. Electrical stimulation of the reticular formation had no effect on the ISPO in the cortex, although the ECoG showed the arousal reaction characteristic of the effect of stimulation. There has also been observed a shift in d.c. potential in the cortex in the form of an aperiodic wave, as described by Arduini et a l . (1957). However, the temporary appearance of the infraslow rhythm has sometimes been recorded in the reticular formation and in the central gray substance. This generally occurred in response to the combined effect of pharmacologic agents and electrical stimulation of brain structures. ISPO's in the reticular formation structures a r e characterized by the brevity of their appearance and by the relatively short latent period before they a r i s e as contrasted with the hypothalamic area. The parameters of the oscillations a r e unstable, although the frequency range is the same as for the hypothalamic area, from 2-3 to 15-20 osc/min. Fig. 51 illustrates an infraslow rhythm of about 11 per min in the reticular formation immediately after stimulation with square pulses for 10 s e c at a frequency of 50 c/s and amplitude of 4 V. In Fig. 51, B, 28 min after electrical stimulation of the reticular formation combined

108

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

Fig. 51. Brief appearance of ISPO in the reticular formation after stimulation by an electrical current with a frequency of 50 c/s (A) and after the intravenous injection of 0.35 mg/kg of phosphacol (B). (A) a = before stimulation; b = immediately afterwards; c = continuation of recordinz b. (B) a = before injection of phosphacol; b = 28 min later.

with administration of a cholinesterase inhibitor (intravenous injection of 0.35 mg/kg of phosphacol), infraslow oscillations appeared among which could be distinguished rhythms of about 1, 4, and 5 osc/min. These ISPO's disappeared 7 min after they occurred. In most cases, however, ISPO's did not appear in the reticular formation even after repeated stimulation. Stimulation of the reticular formation tended to inhibit the ISPO in the cortex, where the frequency and amplitude of the oscillations decreased. Pain, nociceptive stimulation, w a s another agent that evoked ISPO in the reticular formation. Ten experiments were performed in which pain was applied to rabbits by lowering their hind paw into hot water o r stimulating it with an electrical current. ISPO's appeared in the reticular formation immediately after the paw w a s stimulated and disappeared 2-3 min later. The effect was noted in the paraventricular nucleus of the hypothalamus and even more distinctly in the central gray substance and reticular formation. The frequency of the ISPO increased in the cortex and the effect was more persistent. The w a y s in which ISPO's become intensified in the cortex in response to taste or nociceptive stimulation of the paw may well be different. In the former, the hypothalamus plays a larger role, while the reticular formation does so in the latter. ISPO's appear in the reticular formation only briefly and damp off rapidly, whereas they persist for hours and days in the hypothalamus and cortex. Our findings support Anokhin's view (1956) regarding the effect of unconditioned pain stimulation on the rostra1 part of the reticular formation. It is interesting to note, however, that the effect of pain stimulation, judging by the ISPO, remains much longer in the cortex than in the reticular formation, suggesting a possible hormonal origin.

SOME EXAMPLES OF INFRASLOW POTENTIAL OSCILLATIONS

109

Stimulation of the central 'gray substance effects the parameters of the ISPO in the hypothalamus. For example, stimulation of the central gray substance in the walls of the third ventricle intensified the ISPO in the anterior hypothalamus on the contralateral side. The frequency of oscillations was relatively high, 20 p e r min, with an amplitude of 2 mV. The ISPO's became depressed in the cortex after 20 min. In the s a m e rabbit, stimulation of the anterior hypothalamic area intensified the A-rhythm and evoked the B-rhythm in the sensorimotor cortex of both hemispheres while intensifying the more rapid electrical activity, but it failed to evoke ISPO in the central gray substance. Thus, stimulation of the central gray substance may evoke ISPO's in the anterior hypothalamus without their appearing in the particular structure stimulated. EjIc?c 1 of cli lo vpvoi 11 CIZii i e , ad YL' I i ci 1ii i e , n t vop i7i e , ancl plzo spha co 1 01 i ISPO iiz izoiispecific OvoiJi stvitctuves

When we used various pharmacologic agents, it was not our intention to investigate their mechanism of action. The main purpose of the experiments was to gather data on drugs which intensify ISPO in different brain structures. We were guided in our choice of substances by published material on their mechanism of action and points of application. It was shown in Anokhin's laboratory (Agafonov, 1956) that chlorpromazine blocks the effect of nociceptive stimulation. In view of the adrenolytic effect of the drug, it was fair to conclude that this blockade results from the action of chlorpromazine on adrenergic elements (cf. below) in the r o s t r a l p a r t of the reticular formation. Anokhin (1958b) r e g a r d s chlorpromazine as a specific substance capable of blocking only the activating function of the reticular formation caused by adverse environmental factors and the organization of defensive behavior, whereas the reticular formation functions involved in the execution of other reactions ( e . g . , food) are preserved. Chlorpromazine also influences the processes that are reflected in ISPO, the effect being manifested in many p a r t s of the brain. For example, in the experiment on a rabbit shown in Fig. 52, ISPO's were absent in the reticular formation, but two rhythms appeared in the medial preoptic area of the thalamus and in the septa1 area, about 12 and 4 per min. Intramuscular injection of a large dose of chlorpromazine (5 mg/kg) evoked two rhythms in the reticular formation (A) within 3 min: 1-2 p e r min with an amplitude of 3 mV and 1 2 per min with an amplitude of 0.8-1 mV. The effect was relatively brief, disappearing after 25 min. The ISPO's did not change in the nucleus of the septum (B), but they became depressed in the medial preoptic area (C). They were intensified in the cortex 20 min after the injection of chlorpromazine and depressed after 30 min. In another experiment, the injection of chlorpromazine evoked ISPO 1 9 min later with a frequency of about 18 osc/min in the anterior hypothalamus. They became intensified in the central gray substance by the 30th min when they were beginning to disappear in the hypothalamus.

110

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

Fig. 52. Effect of chlorpromazine on ISPO in different subcortical structures, (A) brain stem reticular formation; (B) septum pellucidum; (C) preoptic nucleus. The figures designate the time (in min) after intramuscular injection of 5 mg/kg of chlorpromazine.

In our experiments the injection of chlorpromazine evoked ISPO in the reticular formation, hypothalamic area, and cortex. A feature of the ISPO in the reticular formation is the brevity of their appearance. Reticular neurons differ from hypothalamic neurons in that they react more quickly to chemical agents and regain their original state more quickly. In other words, they are less passive and more stable. Pharmacologic agents may be introduced into the peripheral blood circulation (intravenously, subcutaneously, intramuscularly) and central blood circulation (carotid and vertebral a r t e r i e s ) and ventricular s y s tem of the brain. They can also be applied directly to nervous tissue by local injection and application. To analyze the mechanism of action of pharmacologic substances on the central nervous system, it seems best to compare the results obtained by different methods of administration and to combine them with transection of the brain at different levels and with experiments on isolated cortex. The significance of the method of administration is clearly revealed by a study of the central effects of epinephrine. In 1954, Bonvallet et a l . reported that the arousal reaction in the ECoG in response to intravenous injection of epinephrine is prevented by intercollicular transection of the mesencephalon. Rothballer (1956) found that the structures located below the level of such a transection are related to the activation caused by this hormone. Bonvallet et a2. (1956) concluded from their investigations of single neurons with extracellular microelectrodes that reticular formation has neurons whose activity increases in response to intravenous injection of epinephrine even if all the connections with the other p a r t s of the brain a r e cut (rostrally and caudally). Since the reticular formation is particularly rich in sympathins (Vogt, 1954), Dell and co-workers advanced the idea that the reticular formation

SOME EXAMPLES O F INFRASLOW POTENTIAL OSCILLATIONS

111

has neurons with an adrenergic mechanism of transmission of excitation. There are, however, some other explanations of these phenomena. For example, the reticular formation may be activated not by the direct action of epinephrine, but through changes in cerebral circulation (hypothesis of local vasomotor action). According to a third or metabolic hypothesis, epinephrine influences the reticular neurons through chemical components which a r i s e in the body due to the metabolic action of the hormone; the resultant metabolites effect the reticular neurons. Bonvallet et a l . (1954) rule out the indirect (through chemoreceptors and carotid sinus zone) action of epinephrine because the effect persists after transection at the pontine level and cutting of all the afferent craniocerebral nerve connections, except the olfactory and visual, with denervation of the carotid sinus and aortic reflexogenic zone. The mechanism of action of epinephrine can be better understood from the following facts: intravenous injection (under the conditions mentioned above) causes activation in the ECoG, whereas injection into the carotid artery (pathway to the cerebral cortex), injection into the vertebral artery (pathway to the brain stem), and injection into the ventricular system do not cause such activation (Longo and Silvestrini, 1958; Mantegazzini et al., 1959; Feldberg and Sherwood, 1954). A detailed topographic examination of the localization of the catecholamines in the brain was made by Vogt (1954), who showed that epinephrine and norepinephrine a r e more abundant in the hypothalamic area (cf,Table 11), medial nuclei of the thalamus, mesencephalon, and reticular formation than in the lateral nuclei of the thalamus, cerebral and cerebellar cortex. Intravenous injection of 0.01 mg/kg of epinephrine evokes ISPO's in the central gray substance and changes their parameters chiefly in the hypothalamus (lateral and posterior, in the preoptic area). ISPO's with a frequency of 8-16 per min appeared in the central gray substance 2-5 min after the injection of epinephrine and then became attenuated, but they reappeared about 100 min after injection in recordings from the s a m e electrodes (Fig. 53). The effect w a s absent in the cerebral cortex (two-channel recording). In another experiment, a rhythm of about 18 p e r min appeared in the central gray substance 30-40 min after the injection of epinephrine; B-waves arose in the cortex after 110 min. The frequency of the ISPO in the central gray substance did not coincide with that in the cortex. The intravenous injection of epinephrine likewise intensified the ISPO in the lateral p a r t of the hypothalamus at two periods of time, after 2-20 min and after 100 min (Fig. 54). The ECoG at this time showed an arousal reaction; with respect to the ISPO,the effect was very weak. A second injection of the drug had no effect. Cholinergic substances also change the ISPO in the central gray substance. For example phosphacol (a cholinesterase inhibitor) injected intravenously (0.35 mg/kg) evoked within 2-10 min an ISPO in the cent r a l gray substance with a frequency of .8-15 per min and an amplitude of 0.2-1 mV. The effect was observed in the cerebral cortex after 20

112

ROLE OF CERTAIN SUBCORTICAL STRUCTURES

Fig. 53. Appearance of the infraslow rhythm in the central gray substance 105 min after iiitravenous injection of 0.1 mg/kg of epinephrine. (a) recording from the central gray substance; (b) recording from the sensorimotor cortex, The figures designate the time {in min) after the injection of epinephrine.

Fig. 54. Effect of epinephrine on the ISPO in the lateral part of the hypothalamus

(I and ) in the cerebral cortex (II). The figures designate the time (in min) after the intravenous injection of 0.015 mg/kg of epinephrine. The ISPO's became intensified at two periods of time, after 2-20 min and after 100 min.

rnin in the form of an acceleration of the frequency from about 6 to about 1 2 per min. It is interesting to note that intravenous injection of 10 mg/kg of atropine evoked a biphasic effect quite similar to that of phosphacol. The appearance of the ISPO in the gray substance was recorded 20 min after the injection (18 per min, 1 mV) and after 110 min (10 and 16 per min, 1.5 and 1 mV). The effect of intensified ISPO in the sensorimotor cortex was manifested in the appearance of Aand B-rhythms of high amplitude. Another noteworthy fact is that these agents evoke ISPO's both in the gray substance and in the nuclei of the thalamus, chiefly in the medial and anterior nuclei. For example, very regular ISPO's with a rhythm of 9-10 per min appeared in the dorsomedial thalamus 40-120 min after injection of atropine (Fig. 55). At the same time the ECoG showed the high-amplitude slow waves characteristic of the action of atropine when it is injected parenterally. Potent drugs that play a role in the specific metabolism of the reticular

ACETYLCHOLINE AND INFRASLOW POTENTIAL OSCILLATIONS

113

Fig. 55. ECoG of the sensorimotor cortex (I) and ISPO in the dorsomedial nucleus of the thalamus (11) after intravenous injection of 10 mg/kg of atropine. The figures designate the time (in min) after the injection of atropine.

formation evoke also ISPO's. A distinguishing feature of these potential oscillations in the reticular s t r u c t u r e s is their relative brevity of appearance. Yet, just as in the cerebral cortex and hypothalamus, the effect of ISPO intensification in the reticular formation and central gray substance is manifested at least twice: soon after a drug is administered and a considerable amount of time later. The initial intensification of the ISPO is probably due t o the direct influence of the drug on the neuronal structures. The later intensification is apparently caused by a hormonal factor (secondary) and it reflects the p r o c e s s e s set in motion by the effect of this factor on the metabolism of the neurons.

ACETYLCHOLINE AND ISPO IN THE CEREBRAL CORTEX AND SU BCORTICAL STRUCTURES

The many reviews dealing with the effect of acetylcholine on the central nervous system make it unnecessary to discuss m o r e than a few details of special interest (Mikhelson, 1948; Demin, 1952; Feldberg, 1954a; Dale, 1954; and others).

Distribzttiosz of acetylcholiw m d cliotimstevase in bvain stmctztves The acetylcholine content of the brain ranges in different areas and with different functional states of the animal from 0.5 to 30 mvM per g of f r e s h tissue (McIlwain, 1959), and it is m o r e abundant in the gray substance than in the white. The acetylcholine content of the dog brain d e c r e a s e s in chronically isolated cortex (LissAk et ul., 1952). In the rat brain, it increases during the period of increased excitability preceding a convulsion (Torda, 1953), but d e c r e a s e s under anesthesia (MacIntosh and Oborin, 1953). Nerve s t r u c t u r e s can be excited by the application of acetylcholine

114

ROLE OF CERTAIN SUBCORTICAL STRUCTURES

to the cortex, judging by the change in electrical activity and nature of higher nervous activity. These facts and the antagonism between atropine and phosphacol in their effect on conditioned activity a r e regarded as evidence of the existence of cholinergic synapses in the cerebral cortex and reticular formation (Mikhelson, 1957, 1959). The hydrolysis of acetylcholine catalyzes both the specific enzyme (acetylcholinesterase) and the nonspecific cholinesterase, which inter act with both acetylcholine and other complex esters. For example, the action of the inhibitors that specifically inhibit nonspecific butylcholinesterase, which is probably localized in the neuroglia, is more effective with respect to the cortex than the action on acetylcholinesterase (Desmedt and La Grutta, 1957). From the standpoint of function, these data suggest the existence of a mechanism responsible for "spontaneous" activity. This mechanism is based on tissue production of acetylcholine (local hormone) and on the local effect on neuronal excitability. It has been demonstrated histochemically that cholinesterase activity i s particularly pronounced in the dendrite layers of the cortex and in layer V (Pope, 1952). Analysis of the distribution of cholinesterase in different parts of the neuron, as illustrated by cells in the anterior horn of the rat spinal cord (Giacobini, 1959), shows that it is not evenly distributed. Most activity is found in the cytoplasm and dendrites, least activity in the axons, and 100 times less in the nucleus. Nathan and Aprison (1955) found that 70% of the cholinesterase activity in a rabbit brain homogenate is in the fractions containing large cytoplasmatic particles , mitochondria, and microsomes. According to Toschi (1959), acetylcholinesterase is present in the membranes of microsomes which are 40-60 A thick. The same structures contain vesicles which can be seen on electron micrographs (Palay and Palade, 1955) and are believed to contain acetylcholine. According to data obtained on the rabbit brain (Hebb and Smallman, 1956), 52-690/0 of the total specific cholinesterase is concentrated in the mitochondria. It would seem that the mitochondria contain acetylcholine and nonspecific cholinesterase, while the microsomes contain acetylcholinesterase. The same nerve structure may have cells with very different content of cholinergic substances. For example, acetylcholinesterase activity may be 100 times greater in some cells than in others belonging to the same group in the anterior horn of the rat spinal cord (Giacobini, 1959). Enzymes participating in acetylcholine metabolism a r e present in both cholinergic and noncholinergic neurons (cf. Feldberg and Vogt's classification, 1948). Nonspecific cholinesterase is also found in the glia, Schwann cells of myelinated fibers and vascular tissue (Koelle, 1954). Nonspecific cholinesterase is present in the cerebral arteries of man (Thompson and Thickner , 1953) and in brain tumors of astrocytic orgin (Cavanagh et nl., 1954). Thus, there is evidence of extraneuronal sources of nonspecific cholinesterase in brain tissues. Acetylcholine metabolism varies in different parts of the brain (McIlwain, 1959), as is shown in Table III. The level of acetylcholinesterase activity is particularly high in the

ACETYLCHOLINE AND INFRASLOW POTENTIAL OSCILLATIONS

115

T A B L E 111' ACETYLCHOLINE METABOLISM IN DIFFERENT PARTS OF THE BRAIN

Rate of acetylcholine metabolism pe9- g of f r e s h tissue p e r h) Hydrolysis Sytzthes is Ace t y 1cho 1in Nonspecific esterase e s te ra s e

(m

Part of the brain

~

Cerebral cortex Cerebellar cortex Corpus callosum Caudate nucleus Thalamus Hypothalamus

1.3-3.7 0.09

-

13.0 3.1 2.0

60-100 460 10-15 1900 200-310 190

2-4 0.5 14 2 5 11

supraoptic and paraventricular nuclei of the hypothalamus (Abrahams et aZ., 1957). Application of 8-59 yg of physostigmine to the hypothalamus produces the same effect as electrical stimulation.

Effect of acetylcholine on the injmslow potential oscillations Acetylcholine intensifies the ISPO in many brain structures, the effect bcing triphasic. The amplitude increases soon after the drug is injected into the carotid artery, then after 30-80 min, and, finally, after 2-3 h. The second and third phases also occur when acetylcholine is injected directly into the third ventricle, The frequency of the ISPO decreases from 7 to 5 per min in the cortex 2-10 min after 0.015 mg/kg of acetylcholine is injected into the carotid artery (Fig. 56), while the amplitude and regularity of the waves increases. The effect may be due to the direct action of the substance

I

1 rnin

,

IlmV

Fig. 56. Intensification of ISPO in the cerebral cortex after the injection of 0.015 mg/kg of acetylcholine into the carotid artery. Figures on the left indicate iime (in min) after injection.

on the cellular structures of the cerebral cortex. Reappearances of the effect after 30-80 and 120-150 min a r e apparently caused by acetylcholine

116

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

excitation of the subcortical structures, which results in active humoral factors entering the blood. Injection of acetylcholine into an artery caused the ISPO to accelerate from 7 to 12 per min in isolated cortex rather than to decelerate as in intact cortex. The second phase of the effect was also noted in the strip along with acceleration of the rhythm. In some experiments, acetylcholine was placed in a cannula inserted in the hypothalamic region 10 days before. Again there were two-stage changes in the ISPO of the cortex (Fig. 57). The amplitude and regularity

Fig. 57. Biphasic effect of intensification of ISPO in the lateral nucleus of the hypothalamus after introducing acetylcholine into this nucleus through an inserted cannula. (a) before introduction; (b) 2 min after introduction; (c) 30 min later; (d) 50 min later; (e) 80 min later; ( f ) 90 min later; (g) 100 rnin later; (h) 120 min later; (i) 200 min later.

of the ISPO increased after 50 min and high-amplitude oscillations arose in the ECoG. During the 50-100 min period the ISPO damped off as the ECoG resumed its normal appearance. The effect appeared again after 110-150 min. The ISPO became intensified once more and the ECoG exhibited high-amplitude oscillations , sometimes occurring in s e r i e s . The injection of acetylcholine into the carotid artery produces t r i phasic intensification of the ISPO in the cerebral cortex, whereas injection into the hypothalamic area produces biphasic intensification. The frequency of the ISPO decreases while their amplitude and regularity increase. In isolated s t r i p s of cortex, on the other hand, the frequency after injection into an artery increases. Acetylcholine likewise effects the ISPO in the subcortical structures*. They may a r i s e in the thalamic nuclei which a r e "resistant" to this phenomenon and in the central gray substance. The effect of acetylcholine in the hypothalamic nuclei may

* These experiments were performed jointly with D r .

Koltsova.

ACETYLCHOLINE AND INFRASLOW POTENTIAL OSCILLATIONS

117

be biphasic. For example, in one experiment, ISPO appeared in the late r a l region of the hypothalamus 50 min after the substance was injected intravenously; the frequency quickly reached about 14 per min with an amplitude of 1.5-2 mV. These oscillations damped off after 20 more min. However, within about 100 min of the injection the ISPO again became intensified (12 p e r min) but disappeared after 200 min. The ISPO in the cortex changed almost simultaneously. An example of ISPO intensification in the preoptic region of the hypothalamus is shown in Fig. 58. The rapid onset of the intensification phase

Fig. 58. ISPO in the preoptic region of the hypothalamus after the injection of acetylcholine: {a) before injection; (b) 2 min later; (c) 5 min later; (d) 35 min later; (e) 105-min later.

(within 2 min) is noteworthy. In conclusion we may state that acetylcholine is a very potent agent of intensifying ISPO both in the cortex and in the subcortical structures and brain stem, the effect being manifested in two or three phases. However , injection of the drug does not evoke ISPO in many specific nuclei of the thalamus.

Effect of a cholinesterase inhibitor on infraslow activity in different bra in s tru c ture s The injection of phosphacol, like that of acetylcholine, may intensify the ISPO in the thalamic nuclei, central gray substance, cerebral cortex, and hypothalamus. Intensification of the ISPO in the cerebral cortex occurs simultaneously with the appearance of high-amplitude discharges in the ECoG (Fig. 59) within 25 min of injection of phosphacol. They become intensified once more after 120 min while periodic activity develops in the ECoG. Repeated injections of phosphacol so intensify the effect that the ISPO remains pronounced for 1 - 2 h. The effect appears simultaneously in both hemispheres and in many subcortical structures, even those as resistant (judging from the ISPO)as the reticular formation.

118

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

Fig. 59. Effect of phosphacol on ISPO and the ECoG. (A) ISPO before the experiment; (B) ECoG 25 rnin after intravenous injection of 0.35 m g / k g of phosphacol; (C) ISPO after 80 min; (D) and (E) ISPO and ECoG immediately after a second injection of phosphacol; (F) ISPO on the next day. The I11secTttime marking .applies only to B and E.

The intravenous injection of phosphacol evokes ISPO in the central gray substance within 2-5 min; a brief burst of infraslow oscillations with a frequency of 12-15 per min a r i s e s in this area. A second effect of ISPO intensification can be seen 2-3 h after injection of the drug. Intensification of the ISPO in the hypothalamus after injection of phosphacol occurs within 20-90 min (Fig. 60) and 100-180 min after the

Fig. 60. Effect of phosphacol on ISPO in the sensorimotor cortex (Iand ) in the posterior hypothalamus (u).(I) a = before intravenous injection of 0.35 mg/kg of phosphacol; b = 100 min later; (19a = before injection; b = 90 min after injection.

injection; high-amplitude activity arises simultaneously in the nucleus. Phosphacol, especially in large doses, is the most potent way of intensifying ISPO in the thalamic nuclei. The injection of phosphacol evokes ISPO even in such a "resistant'' structure as the lateral nuclei of the thalamus; the effect once again is

ACETYLCHOLINE AND INFRASLOW P O T E N T U L OSCILLATIONS

119

biphasic, 20-30 min and 120-180 min after injection. The nuclei in which ISPO appeared or became intensified after the injection of acetylcholine and phosphacol* a r e designated on the map with a + sign (Fig. 61).

Fig. 6 1. Effect of phosphacol and acetylcholine on ISPO in different subcortical nuclei. + = intensification of ISPO; - = absence of effect (Cf. Fig. 43 for the names of the nerve structures).

The effect was also noted in the reticular formation, central gray substance, hippocampus, thalamic nuclei, and hypothalamic area. Acetylcholine and phosphacol have been repeatedly shown to exert an effect in the hypothalamic nuclei and in the cerebral cortex in two stages.

* The number of signs inside a nucleus designates the number of leads without regard to the location within the nucleus.

120

ROLE OF CERTAIN SUBCORTICAL STRUCTURES

This applies both to the ISPO and to the appearance of high-amplitude oscillations in the electrocorticogram. Two periods of intensification of the infraslow waves (10-50 and 80-140 min after injection) occur whether the substances a r e injected into the blood or directly into the hypothalamic region. This tends to corroborate the hypothalamic origin of the two phases of intensified infraslow rhythms. The possible appearance, due to stimulation of the hypothalamus, of a new humoral factor that in turn acts on the hypothalamic a r e a has been demonstrated in the case of epinephrine (Tonkikh, 1959). For example, the hypothalamus influences the secretion of epinephrine by the medullary layer of the adrenals while epinephrine influences, directly or indirectly, the hypothalamic function associated with the control of secretion into the blood of several hormonal factors. It is interesting to compare the intervals of ISPO intensification that we observed with the times when the hormones of the posterior lobe of the pituitary flow into the cerebrospinal fluid (after several actions). For instance, 15-20 min after the proximal end of the cervical sympathetic nerve is stimulated, the cerebrospinal fluid of cats is found to contain hormonal substances whose effect is similar to that of the antidiuretic hormone, vasopressin, and oxytocin (Tonkikh, 1959). This is followed by a period in which these hormones a r e not present in the fluid, but 2 h after stimulation, the fluid is once again found to contain a vasoconstricting hormone and, 4 h later, oxytocin. The prolonged latency periods of hormonal action are probably related to the mechanism of interaction of the hypothalamic and pituitary systems. It is fair to assume, therefore, that the remote periods of ISPO intensification in the cortex a r e determined by the times when the hormones are produced. This agrees with the generalized nature of the late effect, which may be manifested in the cortex, hypothalamus, and central gray substance. However, the frequency of the infraslow rhythms differs from structure to structure. The parameter of frequency is apparently determined by the nature of the structure and its response to the appearance of active substances in the medium. Processes reflected in increased frequency of the infraslow rhythm take place in the cortex following the direct action of phosphacol. These processes a r e probably caused by the inhibition of cholinesterase. Two aspects of the mechanism of action of the inhibitor a r e of interest in this connection: (a) the inhibition of nonspecific cholinester ase changes metabolism in the glia, thereby affecting its activity and transmission of active substances to the neurons; (b) the inhibition of cholinesterase in different neuronal compartments induces local changes in the chemical gradients, activating the auto-oscillatory process in the structure. The influence of the anticholinesterase factor will be effective only for a system in which cholinesterase is not distributed diffusely but is included selectively in various microstructures. Therefore, the change in acetylcholine metabolism is not an adequate condition for the appearance of ISPO;it must be combined with the peculiarities of the cell microstructure. This is due to the fact that processes occur rhythmically in certain neurons and that the infraslow rhythm does not arise in other neurons. However , the influences affecting acetylcholine metabolism also

CONCLUSIONS

121

intensify the ISPO in some other brain structures. The reason is that acetylcholine is included in many metabolic processes and, as shown by Demin (1952), it has broad trophic significance for nervous system function. Acetylcholine stimulates the processes of restoration after excitation (Koshtoyants, 1951; Artemyev and Babskii, 1949) and it plays a p a r t in maintaining the resting potential of neuronal membranes. Acetylcholine also influences the oxidation-reduction processes in tissue, thereby increasing oxygen utilization and CO, elimination. Thus, acetylcholine may exert a broad trophic influence on tissues, acting as a catalytic agent on their metabolism and possibly affecting the dynamics of the ISPO in many brain structures. CONCLUSIONS

Brain structures can be divided into two categories according to the mechanism of the ISPO. The ISPO's a r e absent in some structures; processes take place in others that undergo rhythmic changes of an infraslow order. The former, which a r e stable structures from the standpoint of changes in potential, include the reticular formation, central gray substance, and thalamus. The structures characterized by ISPO's a r e found mostly in the hypothalamic area. The cerebral cortex has both stable elements and those in which infraslow rhythmic potential oscillations can occur. The frequency of the infraslow rhythm is about the same in all the structures, but in the hypothalamic nuclei, especially after they are stimulated, the frequency of the oscillations is higher than in the other brain structures. The ISPO-frequency varies with the metabolic characteristics of a given structure and with the effect on the metabolism of those active substances which appear in the fluid surrounding the nerve elements. The increased ISPO-frequency in the hypothalamus may be due to the fact that this a r e a is particularly sensitive to many highly active substances such as substance P, serotonin, histamine, norepinephrine, and others. Increased ISPO-frequency has also been observed in the central gray substance after stimulation and in the area postrema; these regions a r e likewise rich in neurohormones. In addition, the area postrema is regarded as a sensitive chemoreceptor and it contains a large quantity of glia-like elements (cf. Chapter IV). Stimulation of certain subcortical structures may intensify the ISPO in those structures and affect its parameters in the cortex or in other subcortical nuclei. The responses to stimulation of the stable structures and those in which ISPO appear, vary. For example, stimulation of the hypothalamic nuclei results in prolonged intensification of the ISPO in those nuclei and, at the same time, it affects the ISPO in the cortex. On the other hand, stimulation of the thalamic nuclei and reticular formation has no effect elsewhere, judging by the ISPO. The latter appear only rarely in these structures; they a r i s e soon after stimulation and disappear within a few minutes. Their transiency is evidence of the efficiency of the stabilization mechanisms in these structures (cf. Chapter

w.

The ISPO's in different brain structures become intensified after the

122

ROLE O F CERTAIN SUBCORTICAL STRUCTURES

action of nociceptive stimuli. This intensification is the result of a "stress" reaction which mobilizes the neurohormonal function of the brain. It is also possible to identify influences that are specific for different p a r t s of the brain. For example, the intravenous injection of epinephrine is such a specific factor for the central gray substance in evoking ISPO. Its effect on the ISPO in the central gray substance is, in our opinion, indirect. Specific factors that evoke ISPO in the reticular formation and central gray substance include chlorpromazine and agents involved in acetylcholine metabolism (acetylcholine itself, phosphacol, and atropine). Cholinergic substances intensify the ISPO in many brain structures and they are effective with various methods of administration (direct application, injection into systemic or cerebral circulation or into the ventricular system of the brain). Following influences on the cholinergic aspect of metabolism, ISPO's also appear in the nuclei of the thalamus and reticular formation, where they have not been observed in intact animals o r after many other influences. The mechanism of origin of the ISPO after the action of cholinergic substances is partly related to the spatial distribution of cholinesterase, to the fact that it is "built into" the microstructure of the cell. The role of the hormonal function of the brain in the ISPO phenomenon is highlighted by the distant effects of ISPO intensification. Many influences, specific and nonspecific, may cause such intensification in two stages, 20-30 min and 2-3 h after the influences are applied. The two stages a r e manifested in many brain structures: cortex, hypothalamic nuclei, and central gray substances. The two stages follow even after the injection of acetylcholine directly into the third ventricle. It may be that once the hypothalamus is excited, it promotes the appearance of a new factor which, in turn, affects the hypothalamic a r e a some time later. Such a factor evokes ISPO both in the hypothalamus itself, in the central gray substance, and in the cortex. It also acts on isolated strips of cortex. This factor, judging from all the available evide,nce, is a hormone. The following facts may help to elucidate the relationship between hypothalamic activity and ISPO intensification: (1)ISPO in the hypothalamic nuclei may be intensified by the direct application of a cholinesterase inhibitor, by direct stimulation, and by stimulation of the afferent pathways; (2) the excitation traveling along the afferent pathway changes the ISPO (apparently through collaterals from the specific sensory pathway) in the following time sequence: after 10-20 min, the ISPO are intensified in the medial o r posterior hypothalamus; after 20-30 min, in the medial thalamus with no effect in the lateral thalamus; after 30-50 min, in the sensorimotor cortex; (3) stimulation of the hypothalamus increases the frequency of the ISPO in the hypothalamus proper, but decreases it in the cortex. It is more likely that the infraslow rhythm is transmitted from the hypothalamus to the cortex not directly along the nerve pathways, but with the participation of a chemical mediator, as shown by intensification of the ISPO in isolated strips of cortex 20-40 min after stimulation of the hypothalamus.

CONCLUSIONS

123

The transmission of this humoral influence i s accompanied by increased electrical activity in the dorsomedial nucleus of the thalamus. Specific neurosecretory processes presumably take p a r t in manifestation of this effect. These processes a r e associated with the recently described changes in the colloid inclusions in the ascending pathways from the hypothalamus to the thalamus (paraventricular fiber system). Analysis of the ISPO phenomenon raises, therefore, the question of whether there are two forms of hypothalamic neurosecretion: granular secretion proceeding to the pituitary and ascending, possibly colloid, secretion. Thus, the periods of ISPO intensification in the cortex are apparently connected with the times when hormones are produced in the hypothalamic structures. However, the direct effect of certain substances ( e . g . hormones) on cortical structure can also change the parameters of the infraslow rhythm by directly influencing the metabolism of the ''neur oglia-neur on" system. Analysis of the periods of ISPO intensification in the subcortical structures reveals the presence of elements that possess the following three characteristics: (1) they do not respond to transient or random environmental factors, but a r e included after systematic (or extreme) influences; (2) the latent period of their reaction may last scores of minutes; (3) they are involved in regulation for several hours. We relate these characteristics to the properties of the slow control system of the brain.

124 CHAPTER VI

SLOW RHYTHMIC POTENTIAL OSCILLATIONS IN THE LIGHT OF COMPARATIVE PHYSIOLOGICAL DATA

SLOW RHYTHMIC PROCESSES IN CERTAIN ELEMENTARY STRUCTURES

Persistent slow rhythmic processes are found at different stages in animal evolution. Rhythmic activity in a living organism is manifested in two forms: rhythmic movements and rhythmic physicocbemical processes. The latter may be detected chiefly by electrical characteristics. The most elementary manifestations of automatic rhythms a r e the protoplasmatic currents in the simplest organisms and plants. These play a p a r t in the transport of substances. "The most ancient form of manifestation of the major laws governing the evolution of functions the form of cellis increased velocity of the spreading processes to-cell connection within the organism'' (Koshtoyants, 1957). The rhythmic reversible processes of converting protoplasm from sol to gel perform a leading role in such transmission. The relationship between the protoplasmatic currents and various aspects of the enzymatic processes determines their sensitivity to active substances and to electrolytes. Rhythmic changes in potential as well as rhythmic movements have been recorded in plasmodium protoplasm (Kamiya and Abe, 1950). Both rhythms have the same frequency, but there is a constant phase difference between them. Two rhythm frequencies occur at the same time: one half oscillation p e r minute and slower shifts with a period of about an hour (Fig. 62). Artificial halting of the protoplasmatic currents does

...

Fig. 62. Cyclic potential changes in plasmodium protoplasm (Kamiya and Abe, 1950). Y-axis: difference in potentials on both sides of the cell membrane; Xaxis: time in min.

not alter the rhythm of the potential oscillations, a fact indicating that it is somewhat independent on mechanical processes in the protoplasm. These data show that electrical rhythms may appear even at the protoplasmatic level. Frey-Wyssling expressed the view (1948) that protoplasmatic movement is associated with structural changes in such contractile protein

SLOW RHYTHMIC PROCESSES

125

units as actomyosin. The biochemical picture of the rhythm of structural conversions, according to L e (cited ~ by Koshtoyants, 19571, is as follows: "adenosinetriphosphate, by interacting with actomyosine-like bodies, decreases their viscosity and is itself subjected to enzymatic decomposition with the release of energy needed for movement and the formation of adenylic acid; adenylic acid, in turn, by interacting with the same protein bodies, increases their viscosity while rephosphorylation takes place along with the formation of adenosinetriphosphate". The changes that take place in the structure reflect the conversions resulting from the rupture and restoration of specific bonds in the region of the giant molecular structures that a r e caused by enzymatic action. It is interesting to compare the data on rhythmic electrical processes in plasmodium with similar phenomena in plants. Scott (1957) described rhythmic changes in the electrical field of a regular sinusoidal form (with a period of 5-10 per min) on the surface of bean roots. If rootgrowth was inhibited, the oscillations became somewhat less frequent than in the normal growing plant. Their amplitude was highest (2-5 mV) 8 mm from the root tip. In some experiments, another rhythm with a period of 120 min was superimposed on this particular rhythm. The origin of the oscillations w a s localized because the large amplitude of the oscillations occurred in a fairly narrow a r e a of the root. Yet a wave of this rhythmic activity sometimes moved down through the plant. In other cases, oscillations were derived in opposite phases at the same time from the root and from a spot 15 mm away. The author assumes that the root has several sources of oscillation and that there is a mechanism which synchronizes the individual sources. The parameters of these sinusoidal oscillations vary with the salt concentration of the s u r rounding fluid, the physiological state of the plant, and other factors. Thus, rhythmic changes in electrical potential a r e also encountered at the level of plant protoplasm. Slow rhythmic electrical processes a r e likewise associated with the activity of specialized structures responsible for movement but lacking in innervation. For example, rhythmic potential oscillations of sinusoidal form with a frequency of about 2 per min have been recorded between the external and internal surfaces of the cell in ciliated infusorians (Koshtoyants and Kokina, 1957a, b). The electrical rhythm coincides with the rhythm of ciliary movements. The correlation noted between ciliary contraction and cholinesterase activity suggests the existence of a connection between this rhythmic process and acetylcholine metabolism. Acetylcholine is synthesized in many structures possessing rhythmic activity. For example, the frequency of the rhythmic movements of the cilia in mussel gill plates, which have no innervation, is affected by acetylcholine (Bulbring et al., 1953); low concentrations increase the frequency, high concentrations decrease it. The cholinesterase inhibitor physostigmine has the same effect. Epinephrine too accelerates rhythmic ciliary movement. The authors conclude that the constant synthesis and breakdown of acetylcholine in the absence of innervation s e r v e s as a local chemical factor ("local hormone") regulating rhythmic activity. In similar fashion, changes in rhythm frequency with temperature

126

SLOW RHYTHMIC POTENTIAL OSCILLATIONS

change and the rhythm-oxygen content relationship (Gray, 1924) point to a connection between rhythmic activity and the metabolic processes. Rhythmic fluctuations of oxygen tension in pea shoots provide additional evidence of this (Snezhko, 1958). The participation of choline metabolism in the rhythmic processes is determined by the fact that it is "built into" the cell structures and that the latter are not homogeneous. A rhythmic process cannot arise, for example, in a simple, isolated system containing a pure enzyme and a pure substrate dissolved in water or buffer despite a change in reaction rate. This can be shown for a system in which the reaction takes place by the formation of a "substrate-enzyme" complex, provided that this complex breaks down, forming either the original reagents or an enzyme plus the reaction product (P).Such a system, according to Wagner and Mitchell (1958), is described by the equation:

A k Sf + F 4 - S F = F

' k2

k2

f

+P,

where the index f designates a free substance not combined with any other; k l , k , , and k3 are constants of the reaction rates; S is the substrate concentration and F is the enzyme. Let u s consider now whether an oscillatory process can arise in an analogous system. Let there be substance A which can be converted into B at a rate proportional to the concentration of A' (as usual). The products of B can be converted into A by an analogous law and be irreversibly converted into c.

We can then write for the rates of conversion = k2B(t)

- k&(t) .

($)

The rate of the changes of A is regarded by u s as positive if A increases. Therefore, a member corresponding to the conversion of A into B (decrease in A) is taken with a minus sign, while a member corresponding to the reverse conversion (B into A) is taken with a plus sign. By analogy with the calculation that C is not converted back into B:

Let us differentiate the first equation in time:

SLOW RHYTHMIC PROCESSES

Let us substitute here

127

$?from the second equation

and substitute here B(tJ from equation (1)

Let us seek the solution in the form: A whose value is found from the equation

=

1 Xt where X is the parameter

h2 e a t +@?I+k2 + k3)AeAt+ klk3eAt= 0;

The solution of A t is written in the form

where C, and C, a r e arbitrary constants while X, and X, have the sign before the radical in equation (3). It is known that for the solution (4) to be a damped oscillatory process, 1, and X, must be complex values, whereas with undamped oscillation XI and X, must be imaginary. Consequently, the expression under the radical in equation (3) should be less than zero;

and after the transformation we obtain -

This inequality is impossible because a sum of positive values is on the left. If any coefficient, for example k , , were negative, it would mean thatA is converted to B, first, at the rate of k,A (proportional to its own concentration) and, secondly, at the rate of k,B (proportional to the concentration of B). This means that it would be necessary to adopt a scheme whereby B is capable of self-growth at the expense of A at a rate equal to the amount of B already present in, a natural concentration. The incorrectness of inequality (5) shows the impossibility of natural oscillations of concentrationA (and of B at the same time) in the scheme described. The concentration of A in the absence of

128

SLOW RHYTHMIC POTENTIAL OSCILLATIONS

external elements approaches z e r o exponentially with time (1% and X, a r e negative). A rhythmic process in a living structure cannot be activated solely by the presence of enzymes. The intracellular distribution of the enzymes (i.e., the microstructure of the tissue) must play an important role in the origin of the rhythm. Such cell fractions as the nuclei, mitochondria, and microsomes differ in composition of enzymes and they a r e surrounded by a membrane with varying permeability for substances from the external fluid. Under artificial conditions, rhythmic oscillations (auto - oscillatory in character) can be modeled by an apparatus containing a porous membrane. Teorell's model (1958) had a membrane of blown glass carrying a negative charge which separated solutions with different concentrations of sodium chloride. Electro-osmosis was created in the system by passing an electrical current through it. Analysis of the system of equations for this model showed that rhythmic potential and pressure oscillations could arise on the membrane with certain correlations between (a) membrane resistance, (b) distribution of concentration of charged particles thereon, (c) difference in hydrostatic pressures on both sides, (d) potential creating a current through the membrane, (e) rate of electro-osmosis. The period of oscillations depends specifically on the difference in concentrations of the substances on both sides of the membrane. Graphic analysis of the system showed that if at the initial moment the generating point on plane E - P (where E is the applied difference in potentials and P is the difference in pressures) lies in a given limited region, then the resultant oscillations become damped and the system enters a state of balance between the electro-osmotic and hydrostatic pressures. If, however, the initial shock is such that the process begins outside of this region, the oscillations do not become damped. Thus, there can be stable zones in a model with an artificial membrane and all the disturbances occurring inside of this zone fail to give r i s e to an undamped oscillatory process. This physicochemical model suggests to u s that ISPO can a r i s e on a cell membrane when the stable zone decreases, e.g., as a result of a change in the difference in concentration of substances on the membrane. This mechanism is apparently the basis for intensification of the ISPO at a certain level of depolarization of the cell membrane due, for example, to a change in potassium concentration in the external fluid. The auto-oscillatory process s e e m s to be closely related to changes in permeability through the membrane, i.e., to regulation of the internal fluid of the cell. The auto-oscillatory process in unicellular organisms is perhaps a self-regulatory phenomenon, a mechanism involved in adapting metabolic activity to existing environmental conditions. Even at the most primitive stages, infraslow activity reflects the operation of regulatory mechanisms. These primitive regulatory systems have survived in the higher forms of living things just as certain elementary general catalytic systems have passed through the entire complexity of the organic world (Engelhardt, 1961). Like any other control mechanism, this regulation is based on the feedback principle. The principle, according

CLASSIFICATION O F LIVING STRUCTURES

129

to Engelhardt, regulates adenosinetriphosphate metabolism, the inter action of the two main sources of cell energy (respiration and fermentation), etc. Even in unicellular organisms one can distinguish two regulatory s y s tems, rapid and slow. The former, as in infusorians, is manifested in the beating of the individual cilia in response to random environmental factors, to transitory influences. The activity of the slow system is manifested in the existence of an auto-oscillatory process on the cell membrane. Cell excitability, and with it coordination of the strokes of the cilia, changes in time with the rhythmic potential oscillations (Naitoh, 1958). The function of the slow control system is to ensure the stability of the entire organism, to preserve its status among the great variety of external influences. Therefore, the slow control system must contain elements that s t o r e information on earlier experience. LIVING STRUCTURES CLASSIFIED ACCORDING TO THEIR MANIFESTATION OF SLOW RHYTHMIC PROCESSES

Smooth and striated muscles The different muscles, the smooth muscles in particular, a r e characterized by slow rhythmic processes both mechanical and electrical in character. Measurement of the membrane potential of earthworm smooth muscle (Bulbring, 1957) showed that it undergoes oscillatory changes with at least two different rhythms. One has a period of 1 sec, with a spike potential occurring spontaneously at the peak of a slow potential oscillation. The other has a period of about 1-3 min and is equal to the period of mechanical contraction. The magnitude and form of the discharges during these slow potential oscillations (and after artificial polarization of the membrane with a weak current (25 PA) of different direction are different at a maximum and a minimum of the slow wave. The frequency of the slow oscillations recorded from smooth muscles may be altered by a change in concentration of ions in the medium. For example, a decrease in the calcium content of the solution changes in the frequency of the slow potential oscillations in the longitudinal muscle of the intestine from 12 to 9 osc/min. The addition of acetylcholine (2.10*) increases the frequency from 9 to 14 osc/min. The familiar rhythmic activity of the obturator muscle of a bivalve mollusk is regulated by the nearby ganglia. However, Pavlov (1885) noted that rhythmic activity of the posterior obturator muscle sometimes continues even after the ganglia are removed. This observation was confirmed by Barnes (1955), who thinks that rhythmic activity arises in the actual structure of these muscles and that the nerve apparatus simply subjects this rhythm to its control. In muscles capable of highly differentiated activity, "subjection" of the musculature to the nervous system is strengthened and "independent" rhythmic activity in such muscles does not occur. Yet, autorhythmic

130

SLOW RHYTHMIC POTENTIAL OSCILLATIONS

processes may arise in these muscles too under conditions in which the membrane potential of the structure becomes unstable. A change in calcium content of the solution that washes them is a good example of such a condition. Denervation of muscle and actions on its metabolism likewise promote the origination therein of slow rhythmic processes which may be manifested only physicochemically rather than in contractions. Under the influence of a sympathomimetic agent (epinephrine), changes in the electroconducting properties of a tonic (rectus abdominis) muscle of the frog become oscillatory in character (Fig. 63) with a period of - 27400

- 27200 - 27000 - 4U000

I

85.c Bo' 7574,

46W

,

' 1 ICW

Fig. 63. Oscillating character of the changes in electrical parameters of frog muscle after the action of epinephrine (the time of the action is indicated by the arrows and dotted line). (a) and (b) = muscle resistance and capacitance at a currency frequency of 10,000 c/s; (c) and (d) = the same, at a frequency of 180,000 c/s. (I) rectus abdominis muscle; (II) denervated sartorius muscle.

several minutes (Aladjalova, 1950b). This rhythm lasts about an hour. In the nontonic sartorius muscle, after cutting of the motor nerves but with preservation of the sympathetic nerves, the action 'of epinephrine resulted, only 8-9 h after denervation, in the development of rhythmic oscillations in the conducting properties of the muscle. Fig. 63 shows the oscillatory nature of the change in resistance to a current in a frequency range from 10 to 180 kc/s. We regard this (cf. Chapter VIII) as a periodic change in the ion concentration of the substr ate. Orbeli (1923) found that the sympathetic nerves are the carriers of "trophic" influence on the skeletal muscles, causing chemical, physicochemical, and physical changes in muscular tissue. A similar evaluation can be made of the findings on the effects of epinephrine (a sympathomimetic hormone) on the slow rhythmic processes in muscle tissue. The rhythm noted in tonic skeletal muscle is not caused by mechanical processes, whereas in the smooth muscle of mollusks, periodic changes in the conducting properties are associated with contractions (Aladjalova and Mertsalova, 1954). Thus, conditions may be created in muscle tissue that result in the

CLASSIFICATION OF LIVING STRUCTURES

131

occurrence of very slow rhythmic changes that a r e physicochemical in nature. Muscle denervation and the hormonal factor a r e the agents which cause the rhythms to appear. Many examples can be cited from the field of comparative physiology showing that an external stimulus which impairs structural stability may evoke rhythmic processes in a muscle (Koshtoyants, 1957).

Squid giant n e w e fiber The resting potential of a nerve fiber membrane is highly stable. The resting potential of a giant axon membrane kept in good physiological conditions can remain unchanged for several hours. The resting potential of a crustacean nerve membrane (Arvanitaki, 1938; Tobias, 1955) is l e s s stable. Arvanitaki (1938) showed that the potential derived from the surface of a crab nerve fiber may undergo periodic slow oscillations with a period of 0.1-7 sec. These oscillations appear after the nerve is exposed to alcohol or veratrine. The spontaneous discharges characteristic of this nerve invariably occur at the same (in the sense of polarity) peak of a slow oscillation. In other words, there is a correlation between nerve excitability and slow wave phase. The period of this slow rhythm is reduced when the environmental temperature is raised as follows: Temperature ( i n o C)

Period (in sec)

10 18-21 35-40 35-50

2.4-3.6 0.8-1.5 0.34-0.44 0.18-0.28

Sverdlov (1954) recorded on a nerve of a bivalve mollusk regular sinusoidal oscillations of electrical potential with a period of 20-30 sec, which appeared after application of veratrine. It was of undoubted interest to investigate the possibility of deriving very slow rhythmic potential oscillations from the squid giant axon whose membrane potential is extremely stable. Experiments were performed on the giant nonmedullated axon of the squid Omnzatostrephes s banei pacificus, the diameter of which ranged from 100 to 150 p. The difference in potentials was recorded with an intracellular capillary microelectrode filled with a 3 M solution of calcium chloride. Slow rhythmic potentials can be derived by inserting a microelectrode inside the axon, but in doing, so the amplitude of the potentials is 3 -4 times as large as when the recording is made from the outside, although the period remains the same. The resting potential of an axon freshly isolated and kept for 1-2 h in sea water (such an axon is regarded by u s as being in a normal condition) does not change appreciably with time. However, various chemical agents can disturb the stability of this potential and create conditions conducive

132

SLOW RHYTHMIC POTENTIAL OSCILLATIONS

to its undergoing periodic changes. For example, the application of potassium chloride to an adjacent portion of the axon will produce rhythmic oscillations of resting potential in a few minutes (Fig. 64).

-

0'

KCl

- ' 7

16'

25' Alcohol

*

lmin

I

IlmV

Fig. 64. Infraslow resting potential oscillations of the squid giant nerve axon evoked by application of potassium chloride t o an adjacent area. The figures indicate the time after application. Lower recording, after the action of alcohol on the axon.

Three rhythms are apparent: 0.7 osc/min with an amplitude of 0.5-0.8 mV, 8 osc/min with an amplitude of 5-6 mV, and 28 osc/min with an amplitude of 0.1-0.4 mV. These rhythms persist for 20-50 min. Application of alcohol to the axon completely eliminates this slow rhythmic process, which can be compared with that of the ISPO in the brain. The infraslow rhythm does not appear simultaneously all along the axon. For example, it w a s found (with the use of a microelectrode in different regions [Fig. 651) that the picture of the infraslow oscillations is 2

-

3 -

1

5

-

4

6 4

,

1

In d

6

25

, 1 , I . I S ,15 mm

r i r r l 5 4 3 2 1

1 rnin

,

Ilmv

Fig. 65. Topography of slow potential oscillations along the squid giant axon. 1-6 = recordings with corresponding microelectrode positions shown on the scheme below relative to the indifferent electrode (hatched rectangle). Distances between the adjacent points a r e in mm.

different in two regions 1 mm apart, i.e., they a r e present in one but absent in the other. It seems that the infraslow potentials a r i s e in very local portions of the axon and either do not spread at all or do so with a large decrement. Potassium chloride is by no means the only agent that causes rhythmic changes in potential on the axon surface. Fig. 66 illustrates the effect of other agents. The application of a 0.5% solution of physostigmine evoked within 14 min B-waves with a frequency of 2 osc/min and an amplitude of 0.7 mV. They were preceded by a period of spontaneous

CLASSIFICATION O F LIVING STRUCTURES

-‘

14

1 t

2 5’-

rs’a

133

3

1 min

IlrnV

I

-

I--

Fig. 66. Slow resting potential oscillations of the squid giant nerve axon after the action of physostigmine and strychnine. (1) = 0.5% solution of physostigmine: (2) = 1%solution of physostigmine; (3) = 0.05% solution of strychnine. The figures on the left indicate the time after the action.

discharges. A m o r e concentrated solution of the drug (1%)depressed the slow oscillations. Slow periodic oscillations were also evoked by the application of strychnine (0.5% solution). Two rhythms were clearly evident in this experiment: 3 osc/min with an amplitude of 0.8 mV and 32 osc/min with an amplitude of 0.1-0.2 mV at a maximum of these waves. Thus, rhythms of three o r d e r s were found in our experiments: about 1-2 osc/min, 8-9 osc/min, and 20-30 osc/min, which were independent of one another. The ISPO of the axon became intensified 8-12 min after the application of substances with different mechanisms of action and the effect persisted for at least 30 min. The discovery of periodic infraslow changes in axon membrane potential was not unexpected because Arvanitaki (1939a, by c) has described oscillatory changes in giant fiber excitability with a period of 1-2 sec. Periodic slow changes in excitability have also been recorded in the myelinated nerve fiber of the frog (Blair and Erlanger, 1936; Pecher, 1939). Others found that either biochemical agents capable of changing membrane potential o r artificial polarization promote the appearance of periodic infraslow changes in nerve excitability (Monnier, 1952). In Lehmann’s experiments (1937) on the cat phrenic nerve kept for s o m e time in a solution with a high pH, slow periodic potential oscillations appeared in response to stimulation, with b u r s t s of frequent pulses being invariably noted at the c r e s t s of these waves. If a system which is stable under normal conditions should lose its stability and begin to oscillate, it would mean that the system is controlled by a negative feedback mechanism and that an oscillatory proc e s s would arise in the event that one or m o r e elements in the feedback circuit changed. Control by means of a negative feedback cannot be instantaneous, i.e., some delay in the regulatory signals cannot be avoided. If this delay becomes excessive, the system might become unstable and begin to generate. These oscillations would be sinusoidal if the values of the processes included in the feedback were linearly correlated. The period of oscillations could depend on the length of the delay in the feedback circuit. The sequence of processes included in the feedback circuit can be schematically represented as follows. The concentration of

134

SLOW RHYTHMIC P O T E N T I A L OSCILLATIONS

substance A in the axon structure influences the permeability of the membrane for substance B. The passage of B through the membrane influences the membrane resting potential and resistance. If it is further assumed that the properties of the membrane determine, in turn, the rate at which substance A enters or is synthesized just as they influence the inflow (or outflow) of substance B, the feedback will be complete. Oscillations arise if the system becomes unstable. The system becomes unstable in case of impairment of the metabolism of A, which is not distributed uniformly, but is concentrated in individual submicrostructural units. All the factors that influence the r a t e of synthesis and, therefore, the amount of substance A, disrupt the stability of the system for some time. Factors that cause membrane depolarization a r e such influences in this example. The more completely the system is stabilized by the negative feedback, the more quickly will the oscillations become damped. Nondamped oscillations can a r i s e only after prolonged disturbances coming from an external source. The regular form of the oscillations implies the existence of synchronism in the metabolism processes which may be due to the fact that the metabolic activity creates potential and is at the same time regulated in certain elements of the system by the field of this potential. Thus, an electrical field may be an active component of the feedback circuit. Nerve cell and ganglion

Tauc (1955) recorded very slow rhythmic changes in membrane potential of sinusoidal form with intracellular recording from a nerve cell in the abdominal ganglion of the marine mollusk Aplysia. The period of oscillations ranged from 1 to 30 s e c with an amplitude of 2 to 50 mV. The slow rhythm w a s more pronounced after slight artificial depolarization of the membrane. "Spontaneous" discharges with a frequency of 6-15 per s e c and an amplitude of 200 pV were noted at the c r e s t s of these slow waves. In addition to these changes in potential, Tauc described another kind of spontaneous oscillations, "pseudospikes", which have a period of 3050 msec and a frequency of 20 per s e c with an amplitude of 5-15 mV. The "pseudospikes" were invariably recorded during the same phase of the slow potential oscillation. They preceded the "spontaneous" discharges and were regarded by the author as nonspreading local excitation (of electrotonic nature). The "pseudospikes" also arose in response to direct stimulation through an intracellular electrode. The author conjectures that the membrane can be divided into several zones with different levels of excitability, with a subliminal stimulus eliciting a response in only one of the zones. Whether or not excitation spreads from an active zone to adjacent zones depends on the excitability present at the particular moment. The ganglion of APlysiu has two kinds of cells: those possessing "spontaneous" rhythmic activity and those which do not possess it. Their electrogenesis is manifested as follows. (I) Cells which possess rhythmic activity: (a) very slow waves (with

CLASSIFICATION O F LIVING STRUCTURES

135

a period >500 msec); (b) rhythmic local potentials; (c) "spontaneous" discharges. (11) Cells which do not possess rhythmic activity: (a) spreading action potential; (b) damped oscillations. Nerve cells, as already noted, a r e surrounded by neuroglia. The secretory function of the glia in invertebrates i s very marked. Secretory granules visible in neuroglia protoplasm a r e about 0.5 in diameter and capable of entering the intercellular space, e .g.,between the lamellae in the giant fiber membrane (De Robertis and Bennet, 1955). The nerve cells of the earthworm ganglion contain spherical inclusions 400-200 %, and large drops 0.5-1 p in size, which suggests the presence of secretory processes. De Robertis assumes that these cells secrete epinephrine. Secretory nerve cells have also been found in the ganglia of insects (B. and E. S c h a r r e r , 1944) and in the brain of vertebrates (including warm-blooded animals) in the preoptic and paraventricular nuclei of the hypothalamus. Substances extracted from the nerve ganglion affect the rhythmic processes in isolated organs, e.g., they accelerate the r a t e of muscular contraction in larvae of the same species (Kijller, 1948). Neurosecretion may possibly influence other, nonsecretory neurons by stimulating discharges of impulses. Investigations of the secretory, epithelial cells of the salivary gland revealed that there i s a connection between the slow changes in potential and the secretory processes (Lundberg, 1955). The membrane potential changes by 1 0 mV in the direction of depolarization at the time of s e c r e tion. Spike potentials were not recorded from these cells. Hence, local hyperpolarization and depolarization processes may take place in cells that are incapable of generating action potentials. This i s a fact of theoretical significance. The retina of the fish eye (Grundfest, 1958) has cells in which illumination of the eye produces only hyperpolarization of the membrane or hyperpolarization at one band of light waves or depolarization at another. Grundfest thinks that these cells a r e neurosecretory and that the nature of their secretion differs after depolarization or hyperpolarization. Glia cells (tissue culture from a mammalian brain) a r e similar to secretory cells with respect to the electrical processes of the membrane. The resting potential of an astrocyte (Hild et al., 1958) changes in 4-6 s e c in response to electrical stimulation. The cell body contracts after the electrical reaction, although not all parts at the same time (Chang and Hild, 1959). The duration of contraction of most of the cells i s 1.4-3.4 min, the duration of relaxation 6-16 min. The latent period between stimulation and s t a r t of contraction i s 1.5-4 min. Rhythmic activity is a feature not only of astrocytes but also of oligodendroglia cells obtained from the corpus callosum of the rat brain (Lumsden and Pomerat, 1951). The latter contract for 2 rnin and relax for 3 min. Thus, the period of oligodendroglia cell pulsation in a tissue culture i s 5 min and highly constant for each specimen. The astrocytes pulsate l e s s regularly and much more slowly with an average period of 14 min. These cells probably pulsate mechanically in intact

136

SLOW RHYTHMIC POTENTIAL OSCILLATIONS

brain, although the rhythm period may, of course, be different. The significance of this pulsation is unclear, but it appears to be connected with brain metabolism and, perhaps, with the secretory function of the glia. Glia contraction may act as a unique kind of massage influencing the local movement of tissue fluids. Does glia pulsation facilitate the movement of active substances (hormones) for a considerable distance along the nerve fibers ? When we turned from measuring ISPO in isolated cells and fibers to measuring these oscillations in cell complexes (in ganglia), we found them to have a rhythm of the same order. Before going into details about our experimental findings on the dynamics of infraslow changes in potential in ganglia, we must f i r s t discuss briefly the structural characteristics of the brain ganglion of arthropoda, a frequent object of study in our experiments (cf. Fig. 68). This ganglion has a protocerebrum, deuterocerebrum, and tritocerebrum. In the lateral parts of the protocerebrum a r e visual lobes (3) in which electrical activity a r i s e s in response to illumination of the eye (Adrian, 1937). The sensory nerves of the antennae approach the olfactory lobes (A) in the lateral parts of the deuterocerebrum. Fungiform bodies (GT) in the center and anterior part of the protocerebrum a r e connected with the visual and olfactory centers. Numerous fibers pass through the center of the ganglion from both the visual and olfactory lobes and from the fungiform bodies. Here an intricate plexus of fibers is found like axons and dendrites. Potential measurement of the brain ganglion of the crab Cavcinus maenas with plate electrodes 0.5 mm in diameter placed on two opposite sides of the ganglia (cf. Chapter VIII) showed that the potential undergoes periodic oscillations of an infraslow order (Fig. 67). There a r e three kinds of rhythms: "B": 2 osc/min with an amplitude of 2-3 mV; ('A": 7-8 osc/min with an amplitude of 0.7-1 mV; and a more rapid rhythm: 12-14 osc/min with an amplitude of 0.2-0.5 mV. These ISPO's damped off after the animal was placed in the dark. Illumination of the eye intensified them and stimulation of the antenna nerve (olfactory receptor in the crab) increased the ISPO frequency from about 7 to about 11 per min. Chloroform anesthesia completely depressed the ISPO in the ganglion. A more detailed investigation (jointly with Dr. Vetchinkina) w a s made of the ISPO in the brain ganglion of the cricket. A recording microelectrode was placed on the olfactory lobe of the brain ganglion and, in some cases, on the fungiform body. Excitation reaches the olfactory lobe through the sensory nerve of the antenna. Dendritic branches a r e concentrated in the center of the olfactory lobe. In addition, axons proceed through it to the fungiform bodies, to the central body of the ganglion (Fig. 68). Therefore, we thought that we could record the potential chiefly from the dendrite a r e a placing the microelectrode in the center of the olfactory lobe. The indifferent electrode was in contact with the body fluid of the cricket. We sometimes used bipolar recording with a distance of 200 p between the electrodes. The recording was made with a d.c. amplifier and loop oscilloscope. A total of 52 crickets were used in the experiments. In most cases,

CLASSIFICATION O F LIVING STRUCTURES

137

8

1n t n

T

C

f

llmv

4

Y

0 -.A.&---4

I-.

v

- nw-_. 6

,-.

”

-x&%

E

m . e &

_I I -

..

*-

-.^--..

- ,

Fig. 6 7 . Infraslow potential oscillations in the c r a b brain ganglion. (A) after preparation; (B) after 30 min in the dark; (C) a t the time of illumination of the eye; (D) after stimulation of the sensory nerves of the antennae; (E) chloroform ane sthe s ia.

ISPO’s were not detected in the ganglia exposed to various stimulations. Stimulation of the sensory nerve of the antenna immediately evoked (recording started 0.5 s e c after stimulation) wavelike changes in potential lasting 5-10 min (Fig. 68). The parameters of these oscillations are: frequency 8-18 per min with an amplitude (average) of 0.1-1.8 mV; frequency 30-40 p e r min, with an amplitude of 0.1-0.2 mV. These infraslow rhythms a r o s e both in the olfactory lobe on the same side as that on which the receptor was stimulated and in the opposite lobe, reflecting the presence of axon connections between the left and right lobes. Ultraslow oscillations of the same order of frequency and amplitude also appeared in the fungiform body on the side stimulated. A sharp drop in temperature in the measuring chamber halted the infraslow potential oscillations. In 14 instances ISPO’s were found in the olfactory lobes before the nerve w a s stimulated. This may have been due to the stimulation of the ganglion during preparation. These experiments show that potential changes of infraslow order a r i s e in a dense a r e a of dendrite and axon elements upon the arrival of exciting afferent impulses (in addition to the high-frequency potentials known in the literature). We believe that the rapid appearance of the infraslow oscillations is due to the release into the a r e a of the dendrite synapses of active substances which alter the stability of the dendrite potential. Similar results were described by Sverdlov (1954) for the visceral ganglion of the bivalve mollusk isolated with the nerve commissure. Several slow waves (with a period ranging from one to several seconds) of attenuating nature arose in the ganglion after electrical stimulation

138

SLOW RHYTHMIC POTENTIAL OSCILLATIONS GT

I

Fig. 68. Slow potential oscillations in the cricket brain ganglion. Bipolar recording from the olfactory zone (A) and from the fungiform body (GT) (cf.top of the figure). (I) recordings made 1 h after preparation without action; top = recording from A; bottom = from GT; (11) activity A before and 1 sec after stimulation of the olfactory nerve (on the same side); (111) GT activity before and 1 sec after stimulation of the olfactory nerve.

of the commissure. Rapid activity increased both in amplitude and in frequency at the c r e s t s of these waves. Thus, stimulation of the nerve pathways approaching the mollusk ganglion by a single stimulus and stimulation of the cricket nerve ganglion evoked infr aslow oscillations which, however, became quickly damped. The activity of the nerve elements in the ganglion changed in time with the infraslow waves. The chemical medium of the ganglia created.during metabolism is one of the important determinants ,of electrical activity (Koshtoyants et al., 1954). It is apparently associated with the action of hormones, which the ganglion structure of invertebrates possesses in abundance. We assume that there is a hormonal link between nerve stimulation and the occurrence of infraslow potential oscillations. The relatively rapid appearance of ISPO in response to stimulation indicates, in our view,

CONCLUSIONS

139

that neurosecretory processes and elements sensitive to neurohormones a r e present in the ganglion proper. Insects, like crustaceans, have neurosecretory elements within the brain ganglion (in the protocerebrum and tritocerebrum), in the commissural apparatus, and on the surface of the ganglion in the form of masses of cells of different sizes (Koshtoyants, 1957). The brevity of the latent period between nerve stimulation and appearance of ISPO may well be explained by the following scheme: the neuroglia functions as a kind of "chemical depot", which collects and stores physiologically active hormones; excitation is sent from the receptor by an impulse that liberates these active substances. The latter act on the postsynaptic membrane and disrupt its stability, giving rise to infraslow potential oscillations. These oscillations reflect periodic increases and decreases in ganglion excitability which a r e also reflected in the temporary connection between the bursts of rapid activity and a given phase (crest) of the infraslow wave.

CONCLUSIONS

Rhythmic infraslow processes a r e found in the structures of animals at all stages of evolution. They are regulated in the simplest of living things by chemical factors. In unicellular organisms they reflect activity of the homeostatic mechanisms of the cell and regulation of the ionic composition. They also interact with the rapidly acting mechanism of movements,. With the development of living matter there has come into being a division of structures into those in which ISPO can be easily detected and into those in which the process takes place only under very special conditions and is rapidly damped. For example, in the structure of certain smooth muscles auto-oscillatory slow processes occur when functions are normal, but in the muscles intended for more differentiated activity, "subordination" of the muscular structure to the nervous system is strengthened and i t s own oscillatory processes a r e not manifested. However, when the stability of the metabolic processes and membrane potential are altered, the latter begins to undergo slow oscillatory changes. The auto-oscillatory process can also be found in a model that contains an artificial porous membrane. This can be achieved, specifically by changing the concentration of ions on both sides of the membrane. The conditions of the membrane of a nerve fiber, particularly that of a giant axon, are the most stable. Yet external physicochemical and biochemical agents may, by disrupting the stability of the membrane potential (through changes in elements of the negative feedback c i r cuit), evoke slow oscillations of the damped type. Intracellular recording of potential from ganglion nerve cells indicates that there is a difference in "affinity" for the ISPO phenomenon among the nerve cells of the same ganglion. Some do not have it, while in others it occurs spontaneously. There may be cells which respond to stimulation solely by relatively

140

SLOW RHYTHMIC POTENTIAL OSCILLATIONS

slow hyperpolarization o r depolarization changes without generating action potential, whereas other cells are characterized by pulse activity. It was shown that glia cells in a tissue culture have the property of mechanical, rhythmic pulsation. The astrocyte and oligodendrocyte forms of glia have a different rhythm of pulsation. Table IV represents our attempt to classify living tissue according TABLE IV CLASSIFICATION OF LIVING TISSUE ACCORDING TO ITS MANIFESTATION OF ISPO

Structures characterized by the ISPO phenomenon Plasmodium (Kamiya and Abe, 1950) Bean root (Scott, 1957) Unicellular organism (Koshtoyants and Kokina, 1957) Smooth muscle of invertebrates (BUlbring, 1957; Aladjalova and Mertsalova, 1954) Denervated frog skeletal muscles after exposure to hormonal substance (Aladjalova, 1950 a , b) Dendrites Some nerve cells of mollusk ganglion (Tauc, 1955)

Structures in which' ISPO's are absent but may be evoked b y external stimulations Frog skeletal muscles Giant nerve axon, crab and squid (Arvanitaki, 1938; current investigation) Nerve commissure of bivalve mollusk (Sverdlov, 1954) Frog and cat nerves (Lehmann, 1937) Bivalve mollusk ganglion (Sverdlov, 1954)

~

Cricket ganglion (current research)

In the mammalian brain (current research) Hypothalamic nuclei Cerebral cortex Neuroglia cells: oligodendroglia and astrocytes (Lumsden and Pomerat, 1951; Chang and Hild, 1959)

Thalamic nuclei Brain stem reticular formation

to its manifestation of ISPO. One cannot rule out the possibility that in the aggregate of cells those elements which a r e characterized by infraslow activity may influence the excitability of the adjacent cells through an electrical field created by infraslow potential and exerting periodic depolarizing and hyperpolarizing effects on the membranes of the adjacent neurons. The infraslow potential field may be a means of recruiting elements into a single rhythm of cyclic change in excitability.

141 CHAPTER VII

INFRASLOW PROCESSES IN THE BRAIN AS PART OF ITS INTEGRATIVE ACTIVITY

CHANGE IN THE INFRASLOW RHYTHM IN FORMATION O F THE CONDITIONED DEFENSE REFLEX

A major indication of the significance of the infraslow rhythmic processes in the integrative activity of the brain is that they a r e included during certain phases in the formation of temporary connections. The continuous cyclic interaction of the hypothalamus and neocortex through the hippocampus (cf. Chapter V) constitutes one of the substrates of cortical-subcortical integration, This "hippocampal circuit" is found only in mammals and it has evolved as a "mechanism to coordinate the high levels of cortical activity with those integral apparatuses of the emotions which are localized mostly in the subcortical structures" ( h o k h i n , 1958a, p. 309). Since infraslow activity of the brain is associated, in its origin, with the structures involved in the "hippocampal circuit", we thought it would be useful to trace the dynamics of the ISPO in the formation of conditioned reflexes. We (together with Dr. Kol'tsova) recorded ISPO and ECoG in the motor and visual areas of the rabbit cortex and in the deep brain structures, including those forming p a r t of the hippocampal circuit of the same animals. Electrodes were implanted in the hippocampus, thalamus (medial, lateral, and anterior nuclei), and hypothalamus (anterior and posterior) of 7 rabbits. The conditioned defense reflex was formed by combining a light signal with electrical stimulation of a paw. The conditioned reflex (flashes of light at a frequency of 8 c/s) w a s given for 10-15 s e c ; in the last second, reinforcement w a s supplied in the f o r m of an electrical shock applied to a front paw. Some 8-10 combinations at 5 min intervals were presented in each experiment. Paw movement and respiration were recorded on a kymograph. The procedure was carried out every other day on two rabbits. The effect on the ISPO in these cases was weak. W e therefore performed the experiments daily on the other 5 animals; the results were clear-cut. The findings can be most conveniently analyzed by dividing each experiment into four parts: (1) initial period (first 2-3 days) of forming the conditioned reflex; (2) generalization period, from 3 to 7-9 days during which the conditioned reflex could be elicited by several stimuli with occasional closures between signals; (3) fixation period during which there were 5-6 out of 8 possible manifestations of a conditioned reflex; (4) extinction period during which the conditioned stimulus was presented without any reinforcement. During the initial period, infraslow activity became intensified in all a r e a s of the cortex (visual and motor regions of both hemispheres), and the frequency of the rhythm increased from 8 to 11 per min (Fig. 69). At this time the following changes took place in the subcortical structures and hippocampus. A rhythm of about 5 c/s appeared or

142

INFRASLOW PROCESSES IN THE BRAIN

Fig. 69. Change in ISPO in different areas of the rabbit cerebral cortex during formation and extinction of the conditioned defense reflex. (I) left visual cortex; (11) sensorimotor cortex of the left hemisphere; (111) sensorimotor cortex of the right hemisphere. 0 = background before the start of formation of the conditioned reflex; 1 = initial period of formation; 2 = generalization period; 3 = fixation period; 4 = extinction period (the recordings in this and the following figures ,were made before the next presentation of the stimulus).

became intensified in the dorsomedial nucleus of the thalamus (Fig. 70), premammillary area of the hypothalamus (Fig. 71), and hippocampus (Fig. 70). Bursts of more rapid oscillations arose in the supraoptic area. Electrical activity became intensified in the dorsomedial nucleus of the hypothalamus (Fig. 70) and in the anterior thalamus. The activity remained unchanged in the lateral ventral and ventromedial nuclei of the thalamus (Fig. 72). During the generalization period, the amplitude of the ISPO increased markedly in the cortex while the frequency decreased. A rhythm of 3-5 osc/min appeared in some rabbits, "B-waves" (2-3 osc/min) in others. The ISPO's were unstable in thesubcortical structures and indistinct.

CHANGE IN INFRASLOW RHYTHM

143

Fig. 70. Changes in the electrical pattern of different subcortical structures of the rabbit brain during formation and extinction of the conditioned defense reflex. 0 = before the start of formation of the conditioned reflex; 1 = initial period of formation; 2 = generalization period; 3 = fixation period; 4 = extinction period. Recording (A) from the supraoptic nucleus of the hypothalamus; (B) from the dorsomedial nucleus of the hypothalamus; (C) from the dorsomedial nucleus of the thalamus; (D) from the hippocampus.

144

INFRASLOW PROCESSES IN THE BRAIN

Fig. 71. Changes in the electrical pattern of different subcortical structures of the rabbit brain during formation and extinction of the conditioned defense reflex. 0 = before the start of formation of the conditioned reflex; 1 = initial period of formation; 2 = generalization period; 3 = fixation period; 4 = extinction period. Recording: (A) from the anteromedial nucleus of the thalamus; (B) from the .anteroventral nucleus of the thalamus; (C) from the premammillary a r e a of the hypothalamus.

The recording of the dorsomedial and ventromedial nuclei of the thalamus and the hippocampus at this time exhibited synchronous oscillations with a frequency of 4-5 c/s (Fig. 70). Activity decreased at the base of the hypothalamus and in the anteroventral nucleus of the

thalamus.

During the period when the conditioned reflex was being fixed, the ISPO disappeared. Electrical activity decreased in the subcortical

CHANGE IN INFRASLOW RHYTHM

145

Fig. 72. Changes in the electrical pattern of different subcortical structures of the rabbit brain during formation and extinction of the conditioned defense reflex. 0 = before the start of formation of the conditioned reflex; 1 = initial period of formation; 2 = generalization period; 3 = fixation period; 4 = extinction period. Recording (A) from the dorsomedial nucleus of the thalamus; (B) from the ventromedial nucleus of the thalamus; (C) from the lateral ventral nucleus of the thalamus.

146

INFRASLOW PROCESSES IN THE BRAIN

structures and hippocampus. An increased level of activity w a s noted in the anterior nuclei of the thalamus and in the premammillary a r e a of the hypothalamus. The ISPO reintensified in the cortex of both hemispheres the second day after the start of extinction. During the extinction period a distinct rhythm of 4-5 c/s again appeared in the pattern of the medial thalamus and in the posterior hypothalamus. There were no appreciable changes in the other nuclei of these structures during the extinction period. On the fifth day after the s t a r t of extinction, infraslow oscillations in the cortex were like those in an intact animal prior to the formation of conditioned reflexes. There is a period of ISPO intensification in the cerebral cortex during the formation of conditioned reflexes in rabbits. Electrical activity intensifies in the thalamic and hypothalamic areas, as confirmed by the findings of Trofimov et a l . (1958). Our facts show very clearly that the greatest changes in frequency of the ISPO occur during the f i r s t 2 or 3 days after the stimuli are presented. During this time the ISPO's in the cortex increase (from 8 to 12 per min); they a r e paralleled by higher amplitude activity in the electrical pattern of certain nuclei of the hypothalamus and thalamus. After 3-4 days (generalization period) the frequency of the ISPO decreases markedly (to 2 per min), but their amplitude increases. Characteristic synchronous slow oscillations with a frequency of 4-5 c/s continue to be recorded in the record of the medial nuclei of the thalamus, posterior hypothalamus, and hippocampus. The activity is considerably reduced in the lateral and ventral nuclei of the thalamus and in certain hypothalamic structures. Changes in frequency of the ISPO in different phases of the formation of temporary connections apparently signify a different degree of involvement of the slow control system of the brain in these processes. The initial phases in the formation of the conditioned defense reflex a r e presumably accompanied by high activity of the brain structures concerned with control over the hormonal level and making considerable use of the neurosecretory mechanisms of the brain. As the conditioned reflex becomes fixed, the activity of these structures diminishes and the processes in the high-speed systems of the brain gain in importance. Analysis in the light of comparative physiology reveals that at all stages of evolution the metabolic elements involved in the formation of conditioned reflexes are realized by the slow processes (Koshtoyants, 1957). As soon as the organism experiences the need to adapt to environmental conditions, these "lower self -regulatory devices, which functioned well in maintaining certain functions at a constant level, must immediately adapt to the requirements of the particular behavioral act" (Anokhin, 1958a, p. 186). An organized behavioral act undoubtedly involved the neurohormonal complex (whisksupplies the energy needed for this act) in the reaction through the connections of the cerebral cortex with the subcortical structures. Our findings illustrate this view. The intensification of electrical activity in the supraoptic area, paraventricular hypothalamus, and dorsomedial thalamic nucleus during the generalization stages indicates that these structures may be the neural mechanism of generalized hormonal reactions,

SLOW CONTROL SYSTEM O F THE BRAIN

147

Intensification of the ISPO in the formation of conditioned connections and in "stress" reactions (cf. Chapter 111) suggests that the ISPO's r e flect the effect of the higher functions of the cerebral cortex being combined with the autonomic influences through the region of hypothalamic integration; they may well be connected with the emotional sphere. Thus, infraslow activity reflects the processes that play a major role in the integrative activity of the brain. SLOW CONTROL SYSTEM O F THE BRAIN

In studying the control mechanisms of the brain, it is convenient to distinguish the main types of systems, the components and interrelations. Warm-blooded animals have a rapid system and a slow system. The former controls the quick reactions to stimulation, many of which have been studied in detail, e . g . , the orienting reflex. The second evaluates more or less systematically active environmental factors and r e o r ganizes the level of activity s o as to regulate resistance and homeostasis. The slow control system influences the parameters of the rapid system, changing the latter's level of activity. If the brain were to be regarded solely as the organ of adaptation, its task would be to organize the conditions required to keep the variables of the organism within normal limits. In other words, the brain in this case would function as a homeostatic regulator. In this monograph we shall not discuss the properties of the brain and mechanisms as a whole that enable it to function as a regulator. We shall focus only on a few manifestations of such activity, specifically on what we have called the "C-fluctuations" of electrical activity in brain structures. We pointed out in earlier chapters that infraslow brain activity is intensified by certain actions after a long latency period, 30-100 and 120-200 min later. We conjectured that this phenomenon reflects the activity of the slow control system of the brain, one of the functions s e e m s to be not only to automatically adjust themsystemto keeping the internal environment constant but actively to establish a new level of activity. To characterize this system, it was necessary to analyze the electrical activity of certain brain structures whose involvement in the phenomenon of intensification of infraslow activity takes place, we assume, under the conditions of directed functioning of the system. The dynamic properties of this system were investigated by introducing artificial disturbances which disrupted its equilibrium. The subsequent behavior of the system was observed for a considerable period of time. Such a disturbance, which took place in the form of a pulse function and lasted for several seconds, was electrical stimulation of one of the brain structures by pulses of current with a duration of 10 p e c , frequency of 30 and 100 c/s, and amplitude of 4 V. Experiments were performed on 15 nonanesthetized rabbits. Bipolar electrodes were implanted in the sensorimotor cortex of both hemispheres and some parts of the thalamus and hypothalamus. Additionally, a pair of electrodes was implanted in the brain of 5 of the animals for local coagulation of the surrounding area. (This s e r i e s of experiments

148

INFRASLOW PROCESSES IN THE BRAIN

was carried out jointly with Dr. Koltsova). The following subcortical structures were investigated: (a) nuclei of the thalamus: dorsomedial, posterolateral, reticular, anteroventral, pulvinar; (b) a r e a s of the hypothalamus: dorsal, lateral preoptic, supramammillary, ventromedial nucleus; (c) optic chiasma. The electrical activity of the hippocampus was also recorded. In analyzing the results, we were aware that the effect observed after stimulation of one of the subcortical structures is not necessarily connected directly with the activation of the particular structure because other nerve structures are also activated due to numerous connections. For example, in the hypothalamic area, an aftereffect in the form of intensified activity may occur after stimulation of about 30 structures (Niemer et al., 1960). We concentrated on the time intervals in which two stages of intensified infraslow activity would a r i s e in the brain. The following struck u s as particularly significant. When an experiment was performed on an animal for the first time, the agent did not produce the two-stage change in brain activity. Two stages were only evident when the agent was presented on the days preceding the experiment. We shall now describe the repeated experiments of this kind. The "C-fluctuations" of electrical activity that arise after introduction of a disturbance (in the form of the above-described agent) have the following characteristics. Electrical activity is markedly intensified in the stimulated nucleus and certain other nuclei after a latency period of some tens of minutes. This intensification lasts several minutes. The first cycle of "excitation" in these structures is followed by a "rest" period reflecting the original state of the electrical activity. However, 10-20 min later a new wave of "excitation" arises in the same nuclei, despite the absence of additional stimulations; this wave, in turn, gives way to another "rest" period. Such cycles of "excitation" (with a similar type of electrical pattern) a r e periodically repeated as often as five times within 2 h. They are comparable to an oscillatory process of damped nature. Thereupon, temporary, more or less sustained relaxation of activity occurs in all the nuclei. A second stage of "fluctuations" of electrical activity starts after this pause, 2-3 h from the time of the initial disturbance without the presentation of new stimulations. This stage is likewise characterized by several cycles of intensified and inhibited activity, but the electrical pattern is no longer the same as that observed a t the first stage of the "C-fluctuations". The electrical pattern of the first stage has oscillations with a frequency of 5-7 c/s giving way to oscillations with a frequency of 10-16 c/s in the periods of intensified activity. During the second stage these periods a r e marked by high-amplitude and high-frequency discharges with a frequency of up to 28 c/s. The burst or spindle-shaped form of activity takes place during this stage. Cyclic excitations of the second stage last for 1-1.5 h and cease 3-5 h from the first disturbance. However, in some instances the activity of the brain structures halts after the end of the second stage of the oscillatory process at a certain level, which then persists for several days thereafter. It will be observed that recording of the electrical pattern of the

SLOW CONTROL SYSTEM O F THE BRAIN

149 nuclei in these experiments was started only 2-5 min after stimulation s o that the earlier reactions could not be observed. Fig. 73 i s a schematic representation of the phenomenon of "Cfluctuations" in different subcortical structures with examples of the electrograms during the first and second stages of the "fluctuations". The second stage began 130 min after introduction of a disturbance. At this time infraslow activity became intensified in the cortex. This experiment was performed on August 13. On July 14, the reticular nucleus of the thalamus was stimulated for the first time and no I

A

n'

-

-

B

-

'

C

0 58'

30'

70'

. .

.

' ,

122'

1%)

130'

148'

148'

185'

164

180'

189'

175)

2101

192'

2a'

260'

195'

-

2m'

100

200

300

100

2m

300 min

200 IW

0

P" 200 100

'

20

40

60

80

100

120

140

160

180

a 0

220

240

260

280

FO

min

Fig. 73. llC-fluctuationsltof electrical activity in the rabbit hypothalamus and thalamus. Top, electrical pattern of the nuclei: dorsomedial thalamic (A) and dorsal hypothalamic a r e a (B) after stimulation of the latter and after stimulation of the lateral hypothalamic area (C). I = first stage; I1 = second stage. Bottom, graphic representation of the lYluctuationslt of electrical activity in the same brain structures; x = time of recording. The rise in the curve on the graph reflects increase in activity. Y-axis: amplitude of electrical oscillations (in pV); X-axis:time (in min) elapsing since presentation of the disturbance; second stage crosshatched.

150

INFRASLOW PROCESSES IN THE BRAIN

"fluctuations" could be recorded. On July 19, the same nucleus was stimulated again with no "fluctuations". On July 21, the reticular nucleus was stimulated for the third time, resulting in only one stage of "fluctuation" of electrical activity between the 16th and 190th min. The "fluctuations" resembled damped oscillations. During the periods of intensified activity that occurred at the 16th, IlOth, 120th, 130th, 160th, and 190th min, the electrical pattern of the dorsomedial thalamus revealed potential oscillations with a frequency of about 16 c/s. "Fluctuations" occurred in the sensorimotor cortex reciprocally with "fluctuations" in the subcortical structures and consisted of synchronized oscillations alternating with periods of desynchronization. During the period of "fluctuations" ISPO's appeared in the dorsomedial thalamus at the r a t e of 8 per min, in the dorsal hypothalamus 10 per min. No ISPO appeared in the reticular nucleus. A total of 8 "fluctuation" cycles were noted; no burst-like activity arose anywhere. On July 27, after an interruption of a week, the dorsal area of the hypothalamus w a s stimulated. In response "fluctuations" arose in the cortex and 100 min later damped out. Repeated stimulations on the same day evoked "fluctuations" in the dorsomedial nucleus of the thalamus and a hint that the second stage was setting in. The next day, July 28, the effect recurred in the cortex. On August 8, after an interruption of 8 days, the reticular nucleus of the thalamus was stimulated but no "fluctuations" were evident anywhere. On August 10, the reticular nucleus was stimulated again, r e sulting this time in a pronounced effect. The first stage of "fluctuations", which arose after a latent period of 28 min, and the second stage, which a r o s e 120 min after the start of stimulation and was characterized by a burst-like activity in all the structures (except the reticular nucleus of the thalamus), were observed at all the leads. ISPO too failed to appear in the reticular nucleus. The first and second stages included 3 and 4 cycles of excitation, respectively. The "fluctuations" occurred reciprocally in the cortex and subcortical nuclei. On the next day, August 11, "fluctuations" were not evoked by stimulation of the dorsal hypothalamus. On August 13 (cf. Fig. 73), the dorsal a r e a of the hypothalamus was stimulated. The first stage of "fluctuations" arose after 28 min. There was only one stage in the reticular nucleus of the thalamus which developed damped oscillations. A maximum of activity with bursts occurring in this nucleus was noted at the 35th min; ISPO's were recorded there at the same time. The second stage of "fluctuations" occurred after 120 min in the dorsomedial nucleus of the thalamus and in the dorsal a r e a of the hypothalamus. High frequencies initially predominated in the electrical pattern. These were followed by spindle-shaped activity. W e gave this day-by-day account of the experiment in order to show that the two stages of "fluctuations" occur only after systematic actions. The "fluctuation" effect was not observed in experiments on other animals the first day after the introduction of a disturbance. Repeated disturbances produced "fluctuations" similar to damped oscillations. Systematic stimulation gave rise to two stages of "fluctuations".

SLOW CONTROL SYSTEM O F THE BRAIN

151

The reticular nucleus of the thalamus invariably had only one stage and it was of the damped oscillations type. The phenomenon of "C-fluctuations" of electrical activity in our experiments was also evoked by stimu1.ation of the lateral a r e a of the hypothalamus, lateral preoptic area, ventromedial nucleus of the hypothalamus, perichiasmal area, and sensorimotor cortex. The latent period was shortest (from 5 to 28 min) after stimulation of the cortex and perichiasmal area; it lasted 30-50 min after stimulation of the hypothalamic areas. The longest latent period (40-60 min) followed stimulation of the preoptic area. The time when the "fluctuations" ceased also varied from structure to structure. The most prolonged "Cfluctuations" were observed after stimulation of the hypothalamic and neurosecretory areas; they lasted 5 h. Since periodic intensification of the ISPO and electrical activity may also occur in isolated cortex after stimulation of the hypothalamus (cf. Chapter IV), the question a r i s e s whether there is a relationship between the phenomenon of "C-fluctuations" and the mechanism of neurohumoral regulation. Involvement of the pituitary in the origin of the phenomenon is doubtful because coagulation of the gland o r the pathways leading to it from the preoptic areas failed to prevent the "fluctuations" from appearing. Moreover, 1 o r 2 weeks after these actions, the stages of the "fluctuations" became more intense and they appeared in nuclei where they had not been observed prilor to coagulation. For example, prior to coagulation, the second stage of the "fluctuations" was seen only in the lateral preoptic area (after it was stimulated). Within a week of electrocoagulationi of the pituitary (with histological confirmation) stimulation of the lateral preoptic areawzoked intense "C-fluctuations" of activity in the same a r e a as well as in the supramammillary area and ventromedial nucleus of the hypothalamus, the latent periods being 37, 51, and 76 min, respectively. The "fluctuations" had a second stage and they were ended by the occurrence of specific activity with a frequency of about 5 c,/s at all the leads. In another case, two stages of "C-fluctuations" arose in the lateral a r e a of the hypothalamus in response to stimulation thereof, but the phenomenon was absent in the thalami5 nuclei (lateral and dorsomedial). Unilateral coagulation was then carried out in the area of the arcuate nucleus of the hypothalamus, injuring the connections between the infundibulum of the hypophysis and supraoptic area of the hypothalamus. The background activity became intensified 10 days after coagulation in the cortex, lateral and medial nuclei of the thalamus. The lateral hypothalamic area was stimulated against this background (on the coagulated side). Two distinct stages of the "C-fluctuations" arose in response in the hypothalamic area where only the second stage, which appeared 130 min after stimulation, was pronounced. The effect was repeated after 4 days. The intensification of the "fluctuations'' observed in these experiments may have been due to the fact that coagulation by itself is a highly potent stimulus which can activate many regulatory mechanisms of the brain, a peculiar kind of "stressor" capable of causing additional disturbances in the regulatory system.

152

INFRASLOW PROCESSES IN THE BRAIN

By way of verification, we coagulated an area which can scarcely be said to have any direct relationship to the connections of the hypothalamus with the pituitary, namely, the lamina terminalis, the upper end of which abuts the anterior commissure of the hemispheres while the lower end abuts the upper margin of the chiasma. No visible changes were observed in background electrical activity of the brain after 10 days, and the initial stimulation of the dorsal hypothalamus did not evoke "fluctuations". The second stimulation a day later evoked two stages of "fluctuations", which were somewhat more intense than they were before coagulation. On the following days the response to stimulation was weak, but after a 2 day interruption two-stage "fluctuations" were clearly evident in the thalamic and hypothalamic areas. Although coagulation of the hypophyseal t r a c t caused a much sharper difference in the intensity of the "C-fluctuations", coagulation of the lamina terminalis likewise exerted a "facilitating" influence on the appearance of the two stages of "fluctuations". The problem of "C-fluctuations" of electrical activity is still in the experimental stage so that premature conclusions must be avoided. Moreover, in our experiments we observed only one aspect of the phenomenon, i.e., that manifested in changes in the electrical activity of certain areas of the brain. No doubt other processes about which we have no information take place in the intervals between these cycles of excitation. We are now concerned solely with the fact that "fluctuations" may occur reciprocally in the cortex and hypothalamic area. We should like to begin the discussion with a few thoughts about the significance of the "C-fluctuations". The existence of "C-fluctuations" in brain structures and the way the phenomenon is manifested make it possible to analyze the work of the slow control system of the brain by comparing it with a model of a regulatory system whose initial state (equilibrium) is temporarily disrupted and then left alone. The element regulated in this model meanwhile strives to regain the original equilibrium by executing around it a s e r i e s of f r e e damped oscillations. This principle of regulation in its external manifestations resembles the phenomenon of "C-fluctuations" in the first stage of the process, where emergence from the state of homeostasis is indicated by the appearance of a rhythm. The same law of restoration of the equilibrium of the regulatable element is operative in the regulation of certain autonomic functions of the body, e . g . , in a change in the blood sugar level in response to the ingestion of glucose. The former equilibrium or a new one is achieved as a result of a damped oscillatory process (Drishel, 1960). Periodic changes were recorded by Latash (1961) in the EEG of healthy subjects and patients with injuries to the hypothalamus. The changes involved an alternation of phases of synchronization and desynchronization of potential oscillations in response to the subcutaneous injection of low doses of epinephrine. The latent period for the development of desynchronization in the EEG and marked autonomic reactions like the sympathoadrenaline c r i s i s usually lasted 15-60 min, z.e., the effect did not occur in response to the direct action of the epinephrine; rather, according to the author, it was a reflection of regulatory changes

SLOW CONTROL SYSTEM OF THE BRAIN

153

proceeding in opposite directions that might (under optimal conditions) end in damped oscillations, although they sometimes became intensified periodically. EEG "fluctuations" in response to the injection of epinephrine can usefully be compared with "fluctuations" in the electrical pattern of the subcortical structures in our experiments. The principle of damped oscillations does not provide for the second stage of "C-fluctuations". The existence of the second stage supports the assumption that the function of the slow control system of the brain is not solely to adjust automatically the body to the preservation of homeostasis. The regulatory process includes mechanisms that markedly influence the activity of different structures and result in fixation at a new level. By this criterion the slow control system of the brain s h a r e s .the characteristics of a model of an ultrastable system, i.e., a system capable of selectively choosing the necessary values of the parameters, rejecting unstable states, and preserving those which create the stability needed to ensure adaptability to new environmental conditions (Ashby, 1954). The second stage of "C-fluctuations" may well reflect the active reorganization of the system on a new level of operation with allowance for the effect of a new factor (a unique kind of "foresight"). This view is supported by the following facts. Experiments on the same animal showed that introduction of a disturbance did not at first elicit the "fluctuations" reaction. When reintroduced (on the second or third day), the disturbance elicited only the first stage of the "fluctuations" comparable to damped oscillations (one stage). After the experiment was repeated several times a week, the disturbance would result in the complete picture of "C-fluctuations" in two stages. If these "fluctuations" ended in a shift to a new level of activity, the introduction of new disturbances during the new few days had no further effect, i . e . , they did not evoke "fluctuations". However, after an interruption of 7-10 days the entire sequence of the "fluctuations" reappeared, suggesting the existence of "forgetting" processes in the system. The above-described pattern resembles a model in which regulation by the ultrastability principle is included when a disturbance becomes a repeated factor and the system "foresees" further repetitions, choosing the necessary "field" that will ensure the most efficient operation in the new situation. If, however, the organism is subjected to an additional trauma (coagulation of certain brain structures), the second stage becomes pronounced, even, at times, to the point of exaggeration. The change in activity of the structures may then assume a pathological form and lose its regulatory influence. One has the impression that the second stage of activity is caused both by systematic actions and by the p r e s ence of "stress" stimuli in the recent past. The facts examined in this section justify our assumption that the existence of rhythmic processes are even slower than the ISPO and have a period of several tens of minutes. These "C-fluctuations" seem to "regulate" the infraslow activity. Some dynamic characteristics of the system can be found by creating nonstationary processes in the control system by the introduction of artificial disturbances. For example, this approach enabled us to discover that a rhythmic process based on a

154

INFRASLOW PROCESSES IN THE BRAIN

logarithmic law can arise in the posterolateral nucleus of the thalamus under certain conditions. Bursts of impulses a r e generalized in this structure and eventually spread by logarithmic law with repetitions every 20 min (the process is saw-like in nature). The significance of this rhythm is not immediately apparent but, by analogy with radio engineering circuits, the process may act as a time scanner. CONCLUSIONS

Infraslow activity of the cerebral cortex changes after the formation of a conditioned defense reflex in relation to the stage in which temporary connections a r e established. It accelerates at the time when the first closures appear. It increases in amplitude, but slows at the time of generalization and almost disappears as the conditioned reflex is being fixed. Intensification of infraslow activity in the cortex is paralleled by intensification of electrical activity in certain subcortical structures: supraoptic a r e a of the hypothalamus, paraventricular nuclei, dorsomedial nucleus of the thalamus, and hippocampus. W e believe that the activity of these structures is connected with neurosecretory processes and hormonal factors. During fixation of the conditioned reflex, activity in these structures and infraslow activity are weakened simultaneously. Thus, the ISPO's reflect the results of higher cortical functiOns being combined with metabolic (hormonal) factors by the mechanisms responsible for hypothalamic integration. On the other hand, the ISPO mechanisms are closely associated with the activity of the slow control system of the brain. One of the characteristics of this system is that it does not r e a c t to a slight, one-time (accidental) external disturbance. Its reaction to an environmental factor that acts more or less systematically persists for several hours and it may be directed not only at overcoming the changes brought about in the internal environment but at reconstructing the level of activity with due regard for the possible effect of the new factor. The aetivity of the slow control system may be modeled in this respect by a system operating on the ultrastability principle. Another characteristic of the system is the long latency period of the response and the hours-long period of regulation that includes a variety of mechanisms. One way of discovering the control mechanisms may be by tieing into the system a nonstationary transitional process involving a temporary disturbance and then letting the system alone. Under these conditions the electrical activity of certain brain structures undergoes rhythmic "fluctuations" with a period of 20-30 min. There may be one o r two stages of these "fluctuations" lasting 2-3 h each, with a different kind of activity in each. Infraslow activity becomes intensified in the two stages. A rhythmic process occurs during the nonstationary interval with the same period (20-30 min). The discovery of these rhythms will enable u s to investigate the unknown mechanisms of regulation, whose

CONCLUSIONS

155

manifestation under the conditions of an equilibrium we do not know. A preliminary examination suggests that a chemical code underlies the activity of certain links in the control process.

156 CHAPTER VIII

IONIC PROCESSES IN THE CEREBRAL CORTEX INVESTIGATED BY THE METHOD O F CONDUCTIVITY METHOD O F INVESTIGATING CORTICAL IMPEDANCE

Electrolytes and function of the central nervous system The works of Vvedenskii and his school, Verigo (1901), Chagovyets (1898), Lazarev (1916), and others include a mass of factual data as well as a theoretical analysis of the effect of ions on nervous tissue. Recent investigations have disclosed details of the distribution and movement of ions in the process of nerve excitation. This research progress w a s greatly helped by advances made in the method of measuring impedance (Cole and Curtis, 1932-1955). The nature of ion distribution has been most thoroughly studied in single nerve fibers and cells, The potassium concentration within a cell is much higher (almost 40 times) while the sodium and chloride concentrations a r e much lower (7 times) than in the extracellular fluid. This distribution is effected by the metabolic processes, in which phosphorus compounds play a major role. The anion of adenosinetriphosphate may change the fixation of charges on a membrane (Ling, 1957). We also have information on the importance of ATP in the metabolic cycle of excitation. Injection of ATP has been found to change sodium metabolism (Coldwell and Keynes, 1957). High-energy phosphates a r e believed to be essential for the release of sodium from the axon. Using the isotope method and flame spectrophotometry, Keynes (1951) found that 3.7.10"12 M sodium ions per cm2 of surface penetrate into isolated giant axon during each pulse of excitation. At this time the axon loses 4*3*10"12 M potassium ions. Similar correlations were noted in cortical cells during convulsive activity (Colfer and Essex, 1947). The outflow of K? from a cell is not a process characteristic of excitation alone. Isolated nerve in physiologic solution loses K? slowly, the loss being accelerated by low temperature and oxygen deprivation (bixon, 1949a). There a r e conflicting opinions on the physicochemical condition of K within the cell. Some authors (Hodgkin and Keynes, 1953) believe that almost all the intracellular potassium exchanges freely with the extracellular potassium, whereas others (Rothenberg, 1950) find that only 10% of K? is free, the remainder being bound and not exchanging. The discussion is significant in determining the role of K' ions in maintaining resting potential. The loss of K' by the cell necessarily results in depolarization and loss of excitability. On the other hand, an increase in K? concentration of the external solution likewise causes depolarization and loss of excitability. The movement of calcium ions is another controversial subject. Some authors (Heilbrunn, 1957) believe that calcium ions a r e released from the cell when the latter is excited, while others (Keynes and Hodgkin, 1956) have observed the penetration of calcium ions into the squid giant

INVESTIGATION O F CORTICAL IMPEDANCE

157

axon upon excitation. They believe that most intracellular calcium is bound with proteins. Increasing concentration of calcium in the external solution r e s u l t s in a loss of excitability, whereas absence of the element causes spontaneous nerve activity [Fessard, 1936). Hodgkin's explanation (1957) of the phenomenon is that calcium ions a r e adsorbed on the membrane, giving r i s e to local change in the electrical field on the inner side of the membrane without altering the potential between the internal and external solutions. The function of calcium ions is to influence membrane permeability and excitability, not to take direct p a r t in the conduction of an impulse. Cell excitation causes a redistribution of some other ions as well; of these, the behavior of the chloride anion has been studied in great detail. Membrane permeability for many ions increases during the development of excitatory postsynaptic potentials but only for small hydrated ions during the development of inhibitory postsynaptic potentials (Eccles, 1957). Some parts of the membrane ( e . g .the so called "pores"), which have been intensively studied in recent years by electron microscopy (Porter, 1953, 1954; Hodgkin, 1957), play a role in the mechanism of this p e r meability. The sign of an ion charge does not affect its ability to pass through the pores. The nature of the cell response varies with the ability of the cell membrane to permit the passage of various ions under these circumstances : the origination of an action potential spreading without decrement or a local change in potential. The axoplasm of the squid giant axon, according to the calculations of Koechlin (1954), contains 520 mequiv. of cations as compared with only 185 mequiv. of chloride and phosphorus, 75 mequiv. of glutamic and aspartic acid anions, and 20 mequiv. of other organic anions. Some 230 mequiv. of unknown ions remain; these a r e required to balance the cations. The method of measuring impedance has been helpful in obtaining theoretically important facts relating to the mechanism of conduction of excitation through nerves and muscles. The specific resistance of axoplasm and myoplasm has been found to be 2-3 times higher than the resistance of interstitial fluid, whereas the resistance of the surface layer (membrane) is at least ten thousand times higher. These findings were the basis for the theory of the "cable" structure of fibers. The method of measuring impedance has made it possible to investigate ionic mobility in a structure and to determine the temporary relationship between changes in membrane permeability and occurrence of an action potential ( c j . Cole and Curtis, 1938; Chailahyan, 1958) as well as the relationship between membrane permeability and ion concentration in the external medium. Analysis of the significance of ionic changes in the function of central nervous system neurons is complicated by additional factors. The elements of the central nervous system are not all equally sensitive to such shifts. Moreover, electrolytes may effect different processes in the neuron: conduction in the synapses, interaction of the apical dendrite region with the neuron body on the assumption that extracellular currents

158

IONIC PROCESSES I N THE CEREBRAL CORTEX

are involved here (Tasaki et aZ., 1954). The gray substances of the human brain contains (Keynes and Lewis, 1955) the following substances in mmoles per kg of H,O: sodium: 3965; potassium: 108-182; chloride: 37-59; potassium: 1.5-2.8; phosphates: 16-22. The ratio of intracellular to extracellular potassium is 30-35. The content of these substances in brain tissue proper differs from that in the blood and cerebrospinal fluid. For example, there i s 20 times more potassium but 3 times l e s s sodium and chloride in brain tissue than in blood. The total number of cations in brain tissue and blood is 166 (in mmoles per kg of H,O) and 153, respectively as compared with 87 anions in tissue and 132 anions in blood. Study of ion exchange between a nerve cell and its environment h a s revealed that the maintenance of normal duration of cell activity is rendered even more difficult by the accumulation of sodium ions in the cell and decrease in number of potassium ions after each act of excitation. For example, Eccles (1957) calculated with each impulse a stan' d u d motor neuron absorbs about 5.10-L5-10.10-= gequiv. of sodium ions while losing the same number of potassium ions. Roeder (1941) investigated in invertebrate ganglion the relationship between spontaneous activity and concentration of potassium ions in the washing fluid. The activity reaches a maximum with concentrations only slightly above normal. It completely disappears if the amount of potassium is doubled; it is reduced by approximately one-half if the fluid lacks potassium. An increase in potassium concentration causes a rapid increase of activity before it is stabilized at a new high level. The author's explanation of these highly significant findings is that the concentration of potassium does not increase simultaneously in different parts of tissue so that local foci arise with different excitability. According to P r o s s e r (1943), various inhibitors of respiration likewise cause a sharp increase of activity, which is followed by depression. Oxygen insufficiency presumably causes the cells to lose potassium. Similar relationships have been found in the autonomic ganglia of warm-blooded animals (Shevelyova, 1956). The addition of sodium chloride to a perfusion fluid causes spontaneous activity to increase. Weak potassium concentrations above normal intensify reactions to submaximal stimuli while large concentrations paralyze activity. In doing so, the liberation of acetylcholine by preganglionic fibers is not inhibited. This is the reason that Brown and Feldberg (1936) assume that the postsynaptic processes are blocked. The data on potassium dynamics in increased cortical activity are quite contradictory. For example, in Dixon's in vivo experiments (1949b), the amount of potassium increased after a period of convulsive activity. According to other authors (Colfer and Essex, 1947), the quantity of potassium ions decreases in the medium while the quantity of sodium ions increases. It is interesting to compare the findings on peripheral nerves with those relating to the exclusion of sodium ions from a washing solution. Gerard and Libet (1940) recorded the dynamics of the rhythmic activity of isolated frog brain in relation to changes in composition of the washing solution. Electrical activity of the brain decreased in solutions

INVESTIGATION O F CORTICAL IMPEDANCE

159

lacking in sodium, although the osmotic equilibrium was maintained by the use of saccharose. Highly specific reactions have been observed in the cerebral cortex, despite some resemblance in the effect of ions in the solution on brain and peripheral nerve activity. For example, the injection of potassium into the bloodstream or its application to the surface of the cortex in a 1%concentration gives r i s e to the phenomenon of spreading depression of electrical activity and is accompanied by a change in d.c. potential of the cortex in a negative direction (Leg0 and Morison, 1945; Burex and Burexova, 1956). Yet this effect is not elicited by the application of a 3% concentration of sodium chloride. The application of sodium chloride changes the potential in the opposite direction of that caused by the application of potassium chloride. The origin of spreading depression after electrical stimulation of the cerebral cortex is also thought (Grafstein, 1956a) to be caused by the liberation of K? ions at a time preceding or coinciding with the beginning of a negative potential wave. At any rate, potassium ions a r e assigned a specific role in the phenomenon of spj*eadingdepression. The question of antagonism between different ions w a s carefully studied after the detailed investigations of Lazarev (1916). Vorontsov (19241, Vassilyev (19391, and Vinogradov (1952) showed that the blocking of nerve conduction by potassium can be counteracted by calcium. Less attention has been paid to the antagonism between these ions in their effect on the central nervous system. Shtern and Khvoles (1933) found after they injected potassium or calcium chloride into the ventricular system of the brain that they were antagonistic in their effect on blood p r e s s u r e and respiration. Minut-Sorokhtina (1934) discovered that the effect of antagonism between potassium and sodium ions on the frog brain was that the former shortened Tiirk's reflex, whereas the latter lengthened it. Investigation of electrical activity of the cat brain (Bonnet and Bremer, 1937) in an isolated specimen likewise revealed that the addition of potassium initially increases activity and then inhibits it, whereas calcium has the opposite effect. Bureg and Burexova (1956), in experiments on rat brain, discovered the existence of antagonism between potassium ions and magnesium, calcium, strontium, and barium ions in the development of spreading depression of electrical activity. Studying the quantitative relations, the authors concluded that the main characteristics of the effect of potassium ions on peripheral nerve a r e retained even with action on the cerebral cortex. However, the antagonistic effect cannot be attributed solely to the mechanism of action of the ions on membrane permeability of the neuron. The great density and variety of nerve elements in the cerebral cortex render the study of the dynamics of ionic changes in cortical functioning particularly difficult. Therefore, the development of methods that would enable such investigations to be made is well worth while. Impedance of living tissue

Most of the information available on metabolic changes in the composition of brain tissue was obtained by t h e u s e of biochemical research

160

IONIC PROCESSES I N THE CEREBRAL CORTEX

techniques, specifically, by the u s e of isotopes (Palladin, 1953; Engelhardt and Lissovskaya, 1953). These techniques help to throw light on specific exchangeable substances. However, chemical analysis reveals nothing of the dynamics of ionic changes in tissue during different phases of its activity. In the case of nonliving systems such as electrolyte solutions, specific conductivity is a measure of the overall mobility of the free ions, regardless of their qualitative composition. The passage of current in a salt solution is caused by the migration of ions to the electrodes to which voltage from an external source has been applied (in a highpolymer system, there may also be a transfer of electrons from atom to atom, as in a metallic conductor). When the ions reach the electrodes, an excess or deficiency of electrons on the electrode surface is compensated by the transfer of electrons to an ion or from an ion, resulting in neutralization of the ions. The electrons meanwhile switch to the external circuit and create a current there. The rate of ion transfer is determined by several factors, which create resistance to the current. Let there be a field with strength E between the electrodes. Each charge q in the solution is under the influence of the force f = qE, which will move it along the force line of field. The density of the current i through the electrolyte will be defined as i = nfuq, where n is the number of charges per unit of volume and fu is the r a t e of drift acquired by a particle u under the influence of force f. Since the anions and kations differ in mobility (u and uk, respectively), the expression f o r current density in t e r m s of th% number of dissociated molecules per unit volume &Ctakes the form i = q 6CE(u 77

a

+ uk).

This expression holds for the simplest case, for a solution with a weak concentration. If it is more concentrated, the mobility of the ua and uk ions will depend on fields created by the charged particles themselves. For solutions with a weak concentration, resistance will diminish with increasing content of dissolved salts. For solutions with a strong concentration, this law does not hold. Resistance will become a nonlinear function of the number of ions because internal fields will begin to act which depend largely on the magnitude of external voltage. In such cases, the above-mentioned expression for current cannot serve as a quantitative characteristic and some changes will have to be made. The conductivity of living tissues w a s first measured in the middle of the last century by the physicists Peltier, Weber, and Lentz and by the physiologists Dubois-Reymond (1848) and Hermann (1879). The work was continued by Galeotti (cited by Lebedinsky, 19331, Hober (1910), Osterhout (1922), and Lebedinsky (1926-1933). The conductivity of living tissues unexpectedly proved to be far lower than that of the electrolytes in the tissues, suggesting the presence of additional factors restricting the mobility of the ions. These forces may be created by high-molecular complexes whose charges bind ions from the solution. Moreover, when living tissue is included in a d.c. circuit there is a rapid but not exponential decrease in the current in the

INVESTIGATION O F CORTICAL IMPEDANCE

161

circuit. It is thus fair to conclude that ohmic as well as capacitive resistance is found between the electrodes. Deviation from the exponential curve shows that capacitance necessarily varies with the f r e quency of oscillation of the external field and that living tissue has both polarization capacitance and static capacitance. Polarization i s due to the structure of tissue, to the large number of interfaces contained therein. Tissue parameters were studied at different frequencies of voltage applied to the electrodes in connection with investigations of polarization structure (Fricke and Morse, 1925; Cole, 1932; Tarusov, 1939a, b; and others). Total resistance for an alternating current consists of ohmic (R) and reactive (x) resistances and is called complex resistance or impedance (Z). For very low frequencies of a measuring current (10-200 c/s) the impedance of living tissue may be inductive (Cole, 1941, 1955); impedance is capacitive for frequencies of 200 c/s and higher. Capacitive reactance is inversely proportional to the capacitance (C) and frequency of the external field (f). The expression for impedance takes the form 1

Z=R+j2TTfC ' where: j =
where Z is a constant. If only static capacitance i s present, a (polarization factor) equals unity and the element is completely impermeable to ions. If permeability is completely, a = 0. The value of a f o r biological objects (Phylippson, 1930); for frog muscle a = 0.37 (Sapegno, 1930); for s e a urchin egg a = 0.5 (Cole, 1933). Cole, after investigating different suspensions (1928), concluded that cell membranes are the only element that determines the relationship between the impedance of the entire object and the frequency. The total effect from all the membranes in the suspension i s equivalent to the effect of a single impedance element (Z ), which is the same function of frequency as the impedance of the in%vidual cell membrane. Different equivalent circuits consisting of combinations of resistances and capacitances were proposed on the basis of measurement of r e s i s t ance and capacitance of living tissue at low (500 c/s) and high (lo7 c/s) frequencies. Fig. 74 shows an equivalent impedance circuit of an axon.

162

IONIC PROCESSES I N THE CEREBRAL CORTEX

Fig. 74. Equivalent circuit of an axon after Grundfest (1952) and Huxley (1954); below: equivalent circuit of a membrane; R1 and R;?: resistance of intra- and extracellular medium; C , R , and V: capacitance, resistance, and EMF of membrane.

Elements C and R a r e the polarization capacitance of the membrane and its leakage (Grundfest, 1952; Huxley, 1954). RL is axoplasm resistance; R, is resistance of extracellular fluid. Membrane resistance is evaluated from its permeability for K, Na, and C1 ions. The electromotive forces v a r e determined by the relationship between the ions inside and outside the axon. In a resting state, permeability for N a ions is low, and potential at the membrane (v) and membrane resistance a r e determined by the permeability for K ions. During excitation sodium conductivity initially increases with a time constant t = 0.24 msec during which membrane impedance is capacitive. In the next stage of excitation sodium conductivity drops with a time constant z = 8.5 msec and impedance becomes inductive. Potassium conductivity with a time constant 7 = 5.5 msec begins to increase at this time, which also testifies to the inductive nature of the impedance. Specific membrane inductance is Zsm = 0.2 w c m 2 and capacitance Cm = 1 pF/cm2. A s living tissue dies, resistance at low frequencies increases, approaching that at high frequencies (Tarusov, 1935). This may be the result of destruction of the polarization structures o r (from the standpoint of the membrane hypothesis) of the disappearance of semipermeability. This was the basis of Tarusov's suggestion (1939a) that the "degree of tissue viability'' be characterized by the ratio of lowfrequency (lo4 c/s) conductivity to high-frequency (10') conductivity. This ratio is 1 for dead tissue, about 7 for living tissue. The impedance-frequency relationship is usually represented by a cyclic diagram (Fig. 75). Ohmic resistance is plotted on the X axis, reactance on the Y-axis. Points representing different frequencies lie on the circumference. A cyclic diagram can be constructed from three points, based on measurements of R and C at three frequencies. Assuming that a cyclic diagram is in the shape of a regular circle, two values can be determined by interpolation from the graphic of the diagram: resistance of the

INVESTIGATION O F CORTICAL IMPEDANCE

163

Fig. 75. Cyclic frequency diagrams. (a) suspension with polarized particles (after Cole. 1928); (b) frog nerve (after Cole and Curtis, 1936); (c) giant nerve axon (after Cole and Backer, 1941a, b); X-axis: ohmic resistance; Y-axis: reactance ( i n n ) . In figures a and c frequency i s in kc/s, in b frequency is in c/s.

object to direct current Ro and at very high frequency RCQ.However, for living tissue, the points indicating low frequencies do not lie on the circumference, a fact that introduces an e r r o r into the interpolation. By assuming, however, that a biological object can be regarded as a suspension of semiconducting particles in a conducting medium (Cole, 1933), one can determine the resistance and capacitance of the cell membranes included in the suspension. Maxwell offered the following expression for suspensions in which the conducting particles are suspended in the conducting medium in the form of a sphere:

where r is the resistance of the suspension, rl the resistance of the dispersion medium, rz the resistance of the suspended particles, and p the volume concentration of the suspended particles. The formula expresses the relationship between the volume concentration of the suspended particles and the ratio of the resistance of the entire suspension to the resistance of the suspended particles. Maxwell's formula also shows that an insignificant change in volume concentration (caused, for example, by swelling) must bring about an appreciable shift in the value of the resistance of the entire system. Polarization, as Wagner demonstrated (1913), can a r i s e on the boundary between suspended particles and the medium (due to the difference in their conductivities). The uneven ionic velocity of the particles and medium causes an exc e s s of ions of one sign to accumulate on the interface. Diffusion forces will oppose the process of ion accumulation. The degree of polarization will depend on the relationship between the volumes and the conductivities of the medium and particles. Polarization can also be caused by the presence of a particular polarizable membrane around the suspended particles. This idea was the basis for several theoretical investigations in which a biological object w a s regarded simply as a suspension with polarizable particles (Fricke, 1931, 1933). Maxwell's formula (3) can be modified (Cole, 1955) for the case where a particle with radius a is not a homogeneous conductor, but has

164

IONIC PROCESSES IN THE CEREBRAL CORTEX

a core (of cytoplasm) with resistance rin and is surrounded by a thin membrane with resistivity rm. The total resistance of the system will then be :

r

(1 - P b ,

r = rl

+

(f

+ P)

bin+ a) m

r

(1

m

’

(4 )

+ fp)r, + f(l- 9) (rin + a)

where f is a form factor of 1.5 for a sphere. Using this equation, Cole and Curtis (1950) showed that if the resistivity of the cytoplasm of spherical cells is 100fl/cm, the resistance of the membrane is 1000R/cm2 and the radius of the cell is 10- cm; and if these cells a r e suspended in an electrolyte equal to it in volume and with a resistance of 100R/cm, the resistivity of the suspension will be 249.33n/cm. If, however, the membrane resistance were infinite, the resistance of the suspension would be 25011. Thus, only a small part of the current through the suspension goes through the cells. Cole and Curtis, relying on such a theoretical possibility, calculated the resistance and capacitance of the fiber membrane from measurement of the impedance of the muscle and giant nerve fiber of the axon. Their calculations agreed fairly well with the figures obtained by direct measurement of these membrane parameters with a microelectrode implanted inside the cell.

Resistance and capacitance of neuronal membranes The impedance of individual cells is investigated by means of microelectrodes with application of square d.c. pulses. The form and size of the pulse is modified as it passes across the membrane and, by photographing this modification, one can calculate the electrical characteristics of the membrane (Teorell, 1946; Eccles, 1957, BYZOV,1958). In passing through an RC network, the pulse edges change their square front into an exponential one. The time required for the leading edge to reach 1/2.7 of the final value is known as the time constant of the network and is proportional to the product R.C (Fig. 76). Membrane resistance is best measured by a double barreled microelectrode inserted inside the cell. A square pulse is passed through one electrode to an indifferent electrode while the drop in potential on the membrane created by the external current source is recorded through the other electrode (Coombs et al., 1955a). The resistance is calculated by comparing the potential drop at the membrane and at another,known, resistance in s e r i e s with the membrane. This method revealed that the resistivity of motor neuron neuroplasm is somewhat higher than the resistance of the extracellular medium. Membrane resistance, within certain limits, does not depend on the intensity of the measuring current and experimenters were surprised to discover that it remains the same whether the current direction is

INVESTIGATION O F CORTICAL IMPEDANCE

165

Fig, 76. Modification of a s q u a r e pulse (B) in network (A)

depolarizing or hyperpolarizing (Coombs et al., 1955b). The membrane of the cell body differs in this respect from the membrane of the giant axon, whose resistance to a depolarizing current is much lower than to a hyperpolarizing current (Cole and Curtis, 1941). The r e s i s t ance of a motor neuron membrane is about 0.8 Mi2.H the effective s u r face of a neuron is taken as 5.10- cm", the resistivity of the membrane will be 400fi/cm2 (Eccles, 1957). The time constant of the membrane of the giant axon and muscle fiber ranges from 1 to 30 msec (Hodgkin, 1951), while that of the motor neuron (Araki and Otani, 1955) is about 4 msec. However, these findings had to be checked because it was discovered that microelectrodes in intracellular solution have some reactance which in itself distorts the form of the pulse. After carefully compensating this parasitic reactance, Coombs et a l . (1956) found a value of 2.5 msec for the time constant of a motor neuron membrane. Capacitance is determined by dividing the time constant by the r e s i s t ance, 3*10* $; hence the specific capacitance of the membrane is 6 pF/cm2. If the potential on the membrane is 70 mV, the electrical charge on the inner side of the membrane will be about 2.1*10"10 C, which corresponds to an excess of 2 ~ 1 0 "gequiv. ~~ of anions within the cell. This excess of anions within the cell is maintained by the metabolic processes. Table V shows the values which have been determined for the electrical parameters of the membranes of different structures. TABLE V VALUES FOR THE ELECTRICAL PARAMETERS O F MEMBRANES O F DIFFERENT STRUCTURES

Specific Specqic Time resistance capacitance constant R (n/cm2) (C, uF/cm") (t,msec) Frog nerve Squid axon Crab axon Gastropod neuron Motor neuron of frog spinal c o r d

15,000 1,000 5,000 10,000 400

0.001 1.1 1

1.5 6

0.2 1.1

2 150 2.5

166

IONIC PROCESSES IN THE CEREBRAL CORTEX

Tissue impedance measured a t different frequencies

The characteristics of a membrane derived from measurement of its electrical parameters proved to be useful in analyzing the conduction of excitation along an axon by providing detailed information on the dynamics of sodium and potassium movement across the membrane. When one turns from the impedance of individual cells to measurement of the properties of th'e "cell assemblage", it is much more difficult to interpret the results. The possibilities of analysis a r e broadened by comparing the measurements at relatively low and high frequencies (Tarusov, 1939a, b). Ions concentrated at phase boundaries do not participate in the conduction of a low-frequency current" thereby increasing the resistance of the system. I€ low-frequency conductivity changes while highfrequency conductivity remains as before, the phenomenon can be regarded as constituting a change in the conditions of ion concentration at the phase boundaries without any change in the total number of f r e e ions. However, a change in conductivity at low and at high frequencies in the same direction is indicative of an overall change in the mobility and number of electrical charges in tissue. A comparison of the measurements of impedance at low and at high frequencies may provide information about the "structuredness" of tissue. For example, the ratio of resistance at low frequency to resistance at high frequency (ohmic factor) is greater in skeletal than in smooth muscles (Fig. 77). The same thing applies to the capacitive

J-llLmL

:z: ?1e

r

20 10

O Rlf/hf

10 1

2

3

4

5

0

7

8

9

10

11

Xrr/tIr

Fig. 77. Distribution curves of frequency factors for different frog muscles. (I) ohmic factor; (11) capacitive factor. a =' gastric muscle; b = rectus abdominis muscle; c = gastrocnemius muscle; d = sartorius muscle. X-axis: factor values; Y-axis: number of measurements.

factor, to the ratio of reactances at low and at high frequencies. This suggests that the structure of skeletal muscle contains a surface whose charge creates a high concentration of ions in its sphere of influence, These forces a r e weaker in smooth muscle (Aladjalova, 1950a). Investigation of muscle impedance at frequencies ranging from 200 c/s to 200 kc/s (Aladjalova, 1953c; Aladjalova and Maslov, 1957;

INVESTIGATION OF CORTICAL IMPEDANCE

167

Aladjalova and Mertsalova, 1958) revealed anomalously high losses in the low-frequency p a r t of the range. The maximum of losses takes place at frequencies of the order of 5-10 cfs. The a r e a of anomalous losses of high-polymer plastics (Scanavi, 1948) is found in the same range. An anomalously high absorption of energy f r o m an external field at a certain frequency is reflected in the appearance of a maximum on the curve showing the relationship between the factor of dielectrical losses and the frequency of the field applied. The presence of a maximum mebay, 1931) i s evidence of a relaxation mechanism of losses. The losses may be defined as the tangent of the angle (tgS) that complements the angle of the phase difference between the voltage and current up to 90'. The freedom of rotation of the dipole in a complex high-polymer structure depends not only on the size and charge of the polar elements and their connections with the polymer circuit, but also, to a large extent, on the interaction with the surrounding polar elements. A s the temperature r i s e s (with preservation of muscle excitability), the maximum of the angle of losses shifts toward the high frequencies, z.e., the mean length of f r e e movement of the dipole increases as a result of a decrease in the degree of association of the molecules (Fig. 78). Within minor

Fig. 78. Frequency relationship between electrical losses of muscle and the shift therein with temperature change. a = sartorius muscle; b = gastric muscle of frog. X-axis: frequencies, in c / s ; Y-axis: value of tgb. Figures on the right: _temperature, O C.

limits, temperature changes primarily affect the energy of the weaker connections of electrostatic nature. This explains why a temperature change in the curve of losses in smooth muscle is greater than in skeletal muscle, for the structure of the latter possesses more rigid connections (Frank, 1956). These examples illustrate the opportunities available to analyze the results of measuring tissue impedance at different frequencies.

168

IONIC PROCESSES IN THE CEREBRAL CORTEX

The substrate of the cerebral cortex is polyphase with respect to the electrical constants of the system. Cortical cells (nerve and glia) and axons a r e surrounded by a membrane with high resistance (as compared with the internal and external medium). If electrodes a r e introduced into this substrate and they are supplied with voltage from an external source, current density will be greater in the solution surrounding the cell elements, and only a small part of the current will pass through the cells. Therefore, one might assume a pYi0.P.i that cortical impedance is determined chiefly by the properties of the interstitial fluid. However, a large part of the space of the substrate in the upper cortical layers is occupied by dendrite branches. Hence, the condition of the dendrites may affect the magnitude of cortical impedance. Impedance of the visual projection zone of the rabbit cerebral cortex characteristically does not change in response to stimulation of a specific afferent pathway, although high-amplitude potentials develop (Aladjalova, 1953a, 1954a, b, 1955, 1956a). This may be due to the fact that all the neurons in the given area are in a state of activity. The cell bodies of the neurons (Mcllwain, 1959) occupy only 3% of the space, the remainder being distributed between the interstitial fluid, glia, and dendrites. This suggests the advisability of treating with caution attempts to judge the membrane resistance of nerve cells from measurement of cortical impedance. It is safer to assume that impedance of the cortical substrate is determined by the properties of the interstitial fluid, the ionic composition of which is in turn related to the activity of nerve elements of the cortex (the possible role of the dendrites should also be taken into consideration). The specific resistance of the white substance of the brain is approximately If times greater than that of the gray substance. Therefore, the results of measurement of cortical impedance will depend largely on the resistance of the gray substance. Rough measurements of the resistance of various cortical layers (Freygang and Landau, 1955) do not indicate that they differ significantly. When external voltage is applied to plate electrodes placed close together on the surface of a hemisphere [Aladjalova, 1954), force lines of field will travel mostly through the upper layers of the cortex and only in part through the deeper layers. Cortical impedance varies with the frequency of the field applied (Dubikaitis, 1959) in accordance with a law that is common to all living tissues and expressed in equation (2). Cortical impedance i s most conveniently measured in chronic experiments with implanted electrodes, which ensure considerable stability of the initial conditions. Then plate rather than point electrodes should be used to prevent resistance between the electrodes from becoming too high. We used round silver plates 4 mm in diameter placed on a hemisphere 6 mm apart at the centers. Under these conditions, the results of impedance measurement are determined chiefly by the ionic processes in the upper layers of the cortex. Measurements of cortical resistance and capacitance showed that they drop when the frequency of the measuring current is increased (Fig. 79).

INVESTIGATION O F CORTICAL IMPEDANCE

169

Fig. 79. Changes in resistance (solid line) and capacitance (broken line) of the cerebral cortex in relation to the frequency of voltage applied.

To calculate from the measured capacitance the part created by the brain tissue proper, one must subtract the polarization capacitance produced at the electrodes. Accordingly, Dr. Maslov of our laboratory measured the capacitance of physiological solution with the same geometric arrangement of the electrodes as in the case of brain tissue measurement. It turned out that the capacitance of the "electrodessolution'' system at the lowest frequencies was greater than the capacitance of the "electrodes -tissue" system. This difference decreased as the frequency was increased. The capacitance of the "electrodestissue" system then became greater than that of the "electrodes-solution" system. Cole and Backer (1941a, b) noted a similar correlation for nerve fiber, which they ascribed to the inductive nature of impedance at low frequencies. Measurements at one frequency a r e of independent interest. They permit continuous, dynamic observation of ionic changes in the substrate. Measurements should be made at higher frequencies to avoid the frequency range in which the results are considerably complicated by polarization phenomena. We recommend a frequency of 10,000 c/s at which polarization phenomena are quite insignificant, while the sensitivity of the change in impedance value to ionic changes in tissue is fairly high.

Method of simultaneously measuring the ohmic and capacitive components of impedance during a rapidly changing process In most cases impedance is measured by means of measuring bridges. Measurements with bridges require compensation of the unknown

170

IONIC PROCESSES IN THE CEREBRAL CORTEX

parameters of the object with suitable standards. A certain amount of time is expended on compensation, thus complicating the investigation of rapidly changing processes. Therefore, even when a cathode-ray oscillograph and amplifier a r e used instead of the previous bridgebalance indicators (headphones and galvanometer), it is impossible to measure rapidly changing parameters. A partly technical solution of the problem of measuring the parameters of a biological object under dynamic conditions w a s found by Dubuisson (1933), Bozler (1935), and Cole and Curtis (1939). These authors compensated with bridge measuring standards the parameters of the object under study (muscle, nerve) and then stimulated it. Voltage in the diagonal of the bridge changed in proportion to the deviation of the object's impedance from the balanced value. It was possible to trace the change in impedance during a reaction from the dynamics of the c a r r i e r -frequency amplitude on the oscillograph screen. However, this method is not suitable for measuring the ohmic and capacitive components of impedance separately, although it does provide some idea of its total value. Comparatively little attention has been devoted to the problem of measuring the components of impedance separately under dynamic conditions. It generally became of interest to investigators only when they found it necessary to measure capacitance and losses in the capacitor in the presence of a pulse voltage o r during breakdown. Since such measurements did not require a high degree of accuracy, the components of impedance were determined separately by measuring the current and voltage amplitudes, in the object and the phase difference between them. The amplitude and phase were determined simultaneously by means of a very simple Lissajous figure (an ellipse) traced by an electron beam on the screen of a cathode-ray tube (Aseyev, 1954). Any scheme f o r simultaneous measurement of tissue resistance and capacitance during excitation must satisfy certain requirements. The sensitivity of the scheme and range of measurement are determined by the parameters of the tissue, magnitude of change in impedance therein during excitation, and the permissible amount of current passed through the circuit of the object. These values vary with the type of object and size of the electrodes by means of which the object is connected to the circuit. The duration of the intervals in which the electrical parameters must be fixed is determined by the rate of the process under investigation. The frequency range requirements depend on the purpose of the research. A detailed examination of these conditions is beyond the immediate scope of this work, so we shall simply present the final figures that we used. Our scheme had to satisfy the following technical conditions. Resistance and capacitance had to be measured over a frequency range of 200 to 300,000 c / s , The values of resistance ranged from 10 to 10,000R. The capacitances to be measured ranged from 50 to 1,000,000 yuF. The amount of permissible current through the object in the low-frequency p a r t of the range could not exceed 0.1 mA, in the high-frequency p a r t 1.5 mA. The sensitivity of the measurements had to be about 0.1%. The process to be recorded was to last about 0.2 msec.

INVESTIGATION O F CORTICAL IMPEDANCE

171

The technique that we worked out on the basis of the above-mentioned conditions involved still photographing of several ellipses traced by an electron beam on a cathode-ray tube screen (Fig. 80). Each ellipse

Fig. 80. Block diagram of apparatus for separate measurement of ohmic and capacitive components of impedance during a rapid, nonstationary process. Above: s e r i e s of ellipses photographed on a single frame,

characterizes momentary values of the object's capacitance and resistance. The full cycle of geometric transformation of an ellipse reflects the change in electrical components of the period under investigation. Sufficient sensitivity of the measurements is assured by utilizing a zero-balance principle in the scheme, while determination of resistance and capacitance directly from the figures of the ellipse is rendered possible by the u s e of a phase inverter. Ellipses are recorded on stationary film by automatic pulse control of the cathode-ray tube beam. Periodic illumination of the tube and simultaneous shifting of the beam along the screen make it possible to move the figures intermittently and f i x the parameters under investigation at several points in a rapidly developing process. Recording consists of photographing elliptical figures - shifting in relation to one another in time and space - that appear briefly on the screen. The electron beam, kept off the screen before measurement is begun, appears simultaneously with the nonstationary process and traces an ellipse corresponding to the parameters of the object at that particular time. The time at which the beam is directed to the screen is set in relation to the velocity of the nonstationary process in such a way that the image remains virtually unchanged during that time. The next '

172

IONIC PROCESSES I N THE CEREBRAL CORTEX

moment the beam is extinguished for a certain amount of time after which it reappears in the adjacent (in relation to the place it appeared before) part of the screen, where it t r a c e s a figure corresponding to the new momentary values of the object's parameters. Displacement of the image on the screen occurs in synchronism with the course of the nonstationary process. The block diagram of the apparatus is shown in Fig. 80. By successively firing the beam at different points on the screen, one can impress on stationary film a series of images that trace the changes in two parameters during the development of a nonstationary process. The intervals of time between the moments to be fixed and the number thereof a r e determined by the frequency of firing the beam and its rate of horizontal movement. The time intervals between the points to be recorded can be quite short. The conditions that limit the use of the method are determined by the frequency of the alternating current with which the parameters of interest a r e measured and the highest possible rate of photographic recording. The image must r e main virtually motionless for a half cycle of the alternating current because during this time the beam can trace enough of the ellipse to permit calculation of resistance R and capacitance C. However, there must be sufficient exposure to show an image on the film, The required exposure time is determined from the value of the maximum rate of photorecording:

'max

1 F2 (1 + -+)2

m/sec,

where F is the aperture ratio of the objective; r is the magnitude of image reduction; K is a coefficient varying with many variables (K is proportional to the voltage at the accelerating electrodes of the tube, to the number of electrons in the beam, to the sensitivity of the film, to the color of the screen, tint, etc.). Since our scheme provides for image displacement over the screen with stationary film, we can utilize the persistence of the tube phosphor. If moving film were used, the persistence would impair the quality of the photorecording. Under these conditions of photorecording, the time intervals between the points to be fixed can be made brief enough to permit detailed investigation of such a rapid process as solitary excitation of a nerve fiber, which lasts only 1 - 2 msec. For example, with an a.c. frequency of 10,000 c/s, if the major axis of the ellipse on the screen is 2-3 cm and the maximum rate of photorecording is of the order of 400 m/sec, moments of the process can be fixed at 0.5 msec intervals. The number of images that can be recorded on a single frame depends on the size of the screen, maximum rate of photorecording, extent of the changes in the object's parameters during a process, and certain other factors. Thus, it is possible to obtain a clear image of approximately 10-15 Lissajous figures even under comparatively poor conditions. In the case of prolonged observations, these figures can be recorded on motion-picture film by pulse control of the electron beam.

+

INVESTIGATION O F CORTICAL IMPEDANCE

Y I

Yl

173

X

AGIO; A R - 0

At-0 ; A R ) 0

A t * O ; AR+O

Fig. 81. Block diagram of a measuring channel and vector diagram of a phase inverter to measure resistance and capacitance using elliptical figures. Below: transformation of an ellipse after changes in the ohmic and capacitive components of the object's impedance.

Fig. 81 shows the block diagram of a channel that measures the ohmic and capacitive components of impedance by means of elliptical figures. The object (Zob) is connected to an a r m of the measuring bridge and balanced by resistance and capacitance standards (Zst) in parallel. Connected to two other a r m s of the bridge a r e resistances Z1 and Z, (matched in order to simplify the balancing). Voltage from the bridge output after considerable amplification is applied to the cathode tube plates controlling horizontal displacement of the beam. Voltage of the same frequency but shifted in phase enters the plates that control the vertical displacement of the beam through a phase inverter and amplifier. The slope and expansion of the ellipse on the cathode-tube screen vary with the relationship between the amplitudes and phases of the voltages reaching the different deflecting plates. To determine the changes in capacitance and resistance of an object by means of an elliptical figure it is necessary to know the relationship between the voltage at the plates of the cathode-ray tube and the parameters of the object connected to one of the bridge a r m s . Let V be the total voltage at the bridge input, Vout be the voltage at the output, and V, be the voltage at the appropriate arm. Then there will be the following equality for a balanced a.c. bridge: Vout

= vZob - v

'ob

=

'ob

+

-

'st

Zl Z l + z,

'

If the value of Zob is changed by the quantity AZ, voltage in the a r m of

174

IONIC PROCESSES IN THE CEREBRAL CORTEX

At the bridge output there will appear the voltage

Because from the state of equilibrium that existed at the beginning

z ob z,

-

Z,Zst = 0,

then, avoiding the t e r m A Z in the denominator, we obtain:

vout = v

Z,AZ (ZOb + Zst) (Z, +

z,)

*

Since Zob is balanced by a parallel connection of capacitance and r e s i s t ance, the expression for Zob can be presented in the form: 'ob

Z - 1 + joCR

*

Introducing the index "bal" for the balance values of R and C and the index "ub" for those changed, we write after the transformations:

After substituting expression (2) in (1) and considering the gain of the amplifier, we obtain the expression for the voltage at the horizontal deflecting plates:

where a is the constant for the gain, Vt is the peak value of total voltage of frequency g a t the bridge. Since changing resistance Rub is found in both the numerator and the denominator, the function Vhor = f (Rub) must b e hyperbolic in nature. However, for all practical purposes one can choose the resistances of the individual a r m s in such a way that within certain limits the denominator can be regarded as a constant. This requires fulfillment of the inequality aR <
x

= K [a, AR cos b t

+ cp) + %AC sin (at + cp),

(3 1

INVESTIGATION OF CORTICAL IMPEDANCE

175

where the constants K, a and a, depend on the sensitivity of the circuit (sensitivity of beam displacement, amplitude of input voltage, gain, bridge sensitivity, and on the balance value of the object's resistance). If voltage is supplied to the other tube plates so as to create vertical cosine deflection Yvert

= K cos (ot

+ cp),

a sloping ellipse will be formed on the oscillograph screen and the ellipse equation, if it is eliminated from equations (3) and (4), will be:

x = a, my + a,

A C V ~

(5)

If AC = 0, x

= aaRy,i.e., the ellipse is transformed into a straight line (Fig. 82), whose slope is determined by the value of the resistance.

Platinum mhctr

U

suhstancm

surlaca

Fig. 82, Diagram for measuring cortical resistance between microelectrodes with a square pulse (after Freygmg and Landau, 1955).

IfAR=O,

the axis of the ellipse coincides with the vertical line. The abscissa of the point of intersection of the ellipse with the axis determines the magnitude of increment of capacitance AC: X X

y=o

=X = 0

k

a2

AC; AC =-

0

ka2

(6)

To determine the increment of resistance from the elliptical figure,

we find the maximum of variable x by taking the derivative of equation (5) with respect to x = 0,

176

IONIC PROCESSES I N THE CEREBRAL CORTEX X

max

=

K V(a,AR)2 + (u2AC)2.

Solving equation (7) for A R and substituting the value of AC from equation (6),we obtain:

Expression (8) is convenient for calculation in the case of large capacitance changes accompanied by marked expansion of the ellipse. In the case of a narrow ellipse (with low capacitance increments), greater accuracy in determining R can be achieved by measuring the segment xn = n nl (cf. Fig. 81) because it is easy to show that there is the equality xn = Vxamax

- xZo.

Segment xn is determined from the simultaneous solution of ellipse _.

equation (5) and of the major axis of the ellipse (x = ulDRy). Substituting X y = a , n R in equation (5), we get

X

n xn = K a l h R ; A R = - .K a l

(91 Thus, a change in the ohmic component of impedance results in a rotation of the ellipse while a change in the capacitive component results in an expansion, as determined from the simple formulas: X

X

A C = - 0a n d A R = - , n A B in which the constant quantities A and B a r e established by preliminary calibration from the formulas

for the already known changes in AR and AC within a linear function. The data from the phase inversion channel and the difference in sensitivity of the horizontal displacement of the beam a r e taken into account in the constant calibrated coefficients A and B. The phase inverter sewes to establish the required voltage phase on the plates controlling vertical displacement of the beam. The formulas in (lo] can be used for calculation only when the phase is defined by equation (4). Otherwise the principles we s e t forth would be violated and it would be impossible to judge changes in the resistance and capacitance of the object from the ellipse. Phase regulation over a wide range i s achieved by means of a bridge type of phase inverter (Fig. 81). The vector diagram shows that output voltage vector CD is rotated relative to output voltage vector AB by an

INVESTIGATION O F CORTICAL IMPEDANCE

177 angle depending on the parameters of the bridge arms. If both r e s i s t ances in the bridge a r m s are equal and the capacitances a r e equal, phase rotation will obey the law

Particular attention should be paid to the design of the measuring bridge, which is the basic element in the circuit for measuring the impedance components. Each a r m of the bridge may contain any impedance Z. In order to balance the bridge it is necessary that the numerical values of the impedances of the four a r m s constitute a proportion and that the difference in phase shift in two adjoining bridge a r m s be equal. The choice of bridge a r m elements depends on the nature of the object's impedance. We regard the object as capacitance and resistance connected in p a r allel and compare it with capacitance and resistance standards as well as with those connected in parallel in an adjoining arm. It is convenient to connect active resistances in two other symmetrical arms. The technical details of this circuit were published in an earl.ier article (Aladjalova, 1955b). A method of measuring cortical resistance b y means of microelectrodes

The above-described bridge method of measuring impedance is highly sensitive, permitting measurements to be made in a wide range of f r e quencies. However, the method is inapplicable to work with microelectrodes or when it is necessary to measure impedance and potentials s imult aneously Square pulses of current can be used to measure resistance between microelectrodes (Teorell, 1946; BYZOV,1958). The block diagram of the apparatus for measuring cortical resistance is shown in Fig. 82 (Freygang and Landau, 1955). A square pulse of 0.3-0.7 msec duration is applied to the cortex through two electrodes. One consists of several coils of platinum wire resting in a bottomless receptacle placed on top of a hemisphere and filled with physiological solutionl- The other electrode is indifferent. This apparatus is used to obtain a uniform field with parallel force lines. The resistance between these electrodes is 500n; the current during application of a pulse is 3 . 5 mA. Two glass capillary electrodes filled with physiological solution a r e inserted in the cortex. The potential drop in tissue that is measured by these capillary electrodes during the supplying of a square pulse to the other electrodes will be proportional to the resistance of that portion of the cortex which is included between the glas capillaries. The difference in potentials in the resistance of the entire cortical m a s s is 20 mV. Neuron excitability is believed to remain unchanged if the voltage supplied is of brief duration (0.5 msec). The pulse of the voltage supplied and the modified pulse from the cortex are compared on the screen of a cathode ray oscilloscope. This

.

178

IONIC PROCESSES IN THE CEREBRAL CORTEX

method was used to determine the specific resistance of cat cortex in the area of the lateral gyrus, which turned out to be 222XL(with barbiturate narcosis). Limitations of the method are: (a) sensitivity is comparatively low (about 1-3%); (b) the animal must be immobilized; ( c ) a s e r i e s of pulses must be supplied to the substrate, possibly resulting in artificial polarization. The advantage of the method is that it permits highly local determinations of resistance to be made.

ELECTRICAL PARAMETERS OF THE CEREBRAL CORTEX IN A WAKING ANIMAL

The electrical parameters of the cerebral cortex (resistance R and capacitance C) measured at a frequency of 10,000 c / s in a wakeful animal do not remain constant over a period of 24 h. They change slowly with time (Aladjalova, 1955a). Fig. 83 presents examples of changes in the electrical parameters

2500

St0

2/50 SUO 0

I

2

3 Time in h

u

5

6

7

Fig. 83. Changes in the electrical p a r a m e t e r s of the cerebral cortex in four animals under normal conditions. Bottom two curves: - luminal sleep; plotted -on the Y-axis: value of tgb capacitance C , (wF;) X-axis: time in hours.

of rabbit cerebral cortex over 4-7 h. The figures refer to four different animals only exposed to a trepanation operation 7-12 days before the experiment. The figure shows that the electrical parameters of an intact animal undergo oscillatory changes. The period of oscillations differs from animal to animal and it varies with their state and with the effect of the experimental setup on them, ranging from 10 min (in easily excited animals) to 1; h (in passive animals). Such fluctuations in the electrical parameters occur both in the visual region of a hemisphere and in the motor region; the phases of the fluctuations do not coincide. Calculation of the electrical loss angle showed that the change in R and C under normal conditions takes place in such a manner that

ELECTRICAL PARAMETERS O F THE CEREBRAL CORTEX

179

the t g a scarcely changes. Fluctuations of electrical parameters are characteristic only of a wakeful animal and they cease with the onset of physiological or druginduced sleep. For example, after two days of sleep induced by luminal, the magnitude of the electrical parameters remain unchanged for a long time (bottom curve in Fig. 83). Only forced awakening and feeding of the animal quickly causes temporary changes in R and C. Thus, periodic fluctuations of ion conductivitjr in the cerebral cortex accompany wakefulness. To determine whether cerebral excitability changes simultaneously with these fluctuations, we measured impedance and electrical reaction of the cortex to rhythmic stimulation at the same time. We selected an animal in which impedance of the visual cortex underwent periodic, distinct oscillations with a period of about 10 min. The animal's contralateral eye was illuminated at a frequency of 8 flashes per sec. Evoked potentials were recorded at a maximum and a minimum of the curve of visual cortex resistance. When the resistance of the cortex increased, the evoked potentials were isorhythmic with the stimulation; when the resistance dropped, no reaction to the light was observed. Thus, as cortical resistance increases, its capacity for assimilation of the rhythm of light stimulations is higher than when it decreases. Fluctuations of the ionic processes in the substrate of the cortex are evidently connected with fluctuations of excitability.

Change in cortical resistance induced b y altering the oxygen supply and carbohydrate metabo lism The brain utilizes about 25% of the oxygen entering the body. Different parts of the brain and even different parts of the neuron react in different ways to anoxia. The dendrites are the most sensitive, being the first to lose excitability. The axons remain excitable for a long time even in a nitrogen atmosphere. Impairment of the oxygen supply of the brain has a perceptible effect on the electrical parameters of the cerebral cortex. Asphyxiation of an animal by compression of the trachea increases cortical resistance by 3-4%. The changes are reversible after resumption of breathing (Fig.

84).

Cerebral hypoxia induced by applying a ligature to the carotid arteries results in increased resistance for 10 min. In time, even though the ligature is kept on, resistance begins to drop, probably due to inclusion of collateral blood circulation compensating for the deficiency of oxygen. Combining ligation with asphyxia may lead to death, which is preceded by a 12% increase of resistance (Fig. 85). Epinephrine and insulin a r e agents known to effect metabolism (chiefly, carbohydrate). Ten min after intravenous injection of 0.03 mg/kg of epinephrine, resistance of the motor cortex begins to increase, achieving a maximum of 20%, while capacitance decreases by 17%. Injection of a lethal dose (0.5 mg/kg intravenously) likewise results within 10 min in increased resistance, this time by 30%, which continues

180

IONIC PROCESSES IN THE CEREBRAL CORTEX

R

Fig. 84. Changes in the electrical parameters of the cerebral cortex in asphyxia. R = resistance (n); C = capacitance (W);t = compression; 4 =freeing of trachea.

Time in h

Fig. 85. Effect of impaired oxygen uptake on resistance R and capacitance C of the brain. (a) ligation of carotid arteries; (b) compression of trachea; (c) period of death.

to increase rapidly in the terminal stage. It will be noted that the electrical parameters of the cerebral cortex scarcely change during the first 10 min after injection, when blood pressure is known to r i s e sharply (cf. following section). Resistance increases while blood pressure is returning to normal. Insulin hypoglycemia increases cortical resistance in the phase p r e ceding insulin convulsions with an almost simultaneous decrease in the sugar level of the blood. Injection of glucose halts the convulsions as electrical resistance of the cortex begins to diminish. Thus, impairment of the metabolic processes following anoxia and the administration of agents that affect carbohydrate metabolism alters the electrical parameters of the cortex by 5-30%. In hypoxia, there is a marked increase in the amount of lactic acid and inorganic phosphates in the brain, but the phosphocreatine content decreases. We have no reason to link the increase of cortical resistance in asphyxia directly to these changes. The alteration of resistance may be the result of secondary impairments caused, for example, by

ELECTRICAL PARAMETERS O F THE CEREBRAL CORTEX

181

a change in enzyme activity or volume (by swelling) of individual tissue components, etc.

Effect of cerebral hyperemia and level of intracranial pressure on the electrical parameters of the cortex

To analyze the factors for the changes in cortical impedance, it was necessary to determine -the extent to which changes in dilation of the blood vessels and intracranial pressure could influence the electrical parameters. Accordingly, we designed experiments involving an artificial change in the rate of blood flow in the cerebral vessels with minimum influence on the metabolic level. The r a t e of blood flow in our experiments was slowed by artificial elevation of intra-alveolar pressure. Compression of the lung capillaries makes it difficult for blood to flow from the right ventricle to the left atrium s o that venous stasis develops in the systemic circulation and the rate of blood flow in the cerebral vessels slows. The sharp drop in arterial pressure noted during the elevation of intrapulmonary pressure also contributes to this slowing of the r a t e of cerebral blood flow. For example, whev air w a s forced into the lungs of an animal, its blood pressure in the carotid artery fell from 140 to 60 mm Hg. Electrical resistance of the cortex during this time increased by only 0.4% while capacitance decreased by 0.2%. The passive oufflow of blood from the cerebral vessels that takes place when an animal is held in a vertical position head up has only an insignificant effect on the electrical parameters during the first few minutes (0.1-0.25%). The blood supply of the brain i s soon impaired as a result of the developing collapse. This period is associated with a 3-8% change in impedance. Blood p r e s s u r e was rapidly elevated by the intravenous injection of epinephrine. This caused a constriction of the peripheral vessels, which was immediately followed by acceleration of the blood flow in the c e r ebral vessels. If the initial blood p r e s s u r e is lower, the vascular effect resulting from the injection of epinephrine is much stronger. Consequently, we performed experiments with an initial blood pressure of 105 mm Hg. After epinephrine w a s injected, it r o s e to 220 mm Hg (Fig. 86). As long as the p r e s s u r e remained at a high level, resistance tended to decrease, but capacitance increased by 0.5-0.8% However, when the p r e s s u r e began to drop, resistance increased, this time probably as a result of metabolic changes (due to the effect of epinephrine on carbohydrate metabolism). That is why the return of the pressure to the original level w a s not paralleled by a restoration of electrical r e s i s t ance, which continued to grow. Our findings suggest that a change in diameter of the cerebral vessels in itself has much less influence on the electrical parameters of the cerebral structures than do the processes responsible for metabolic changes. For example, during asphyxia, when blood pressure rose comparatively little (from 138 to 152 mm Hg), cortical resistance increased

182

IONIC PROCESSES I N THE CEREBRAL CORTEX

mm Hg I

Rn

180

R

430

160

140

410

120

100

300

80 60

0

1

2

3

4 5 6 Time in min

7

8

0

10

370

Fig. 86. Changes in cortical resistance (R) and blood pressure (d) after the injection of epinephrine.

considerably, by 5%. This increase could not have been caused by increased blood pressure o r accelerated blood flow because such changes in hemodynamics reduce rather than intensify cortical resistance. Kedrov and Naumenko (1954) used the method of conductivity to study cerebral blood circulation. According to their observations, changes in vascular dilatation in the brain a r e caused by relatively minor changes in cortical impedance of l e s s than 0.5%. Our own experiments provide confirmation of the phenomenon. On the other hand, shifts in the metabolic processes change the conducting properties of the cortex by 5-20%. Changes of this magnitude cannot be the result of mechanical constriction or vascular dilatation; rather, they a r e caused by ionic changes in the substrate of the cerebral cortex. It was also necessary to determine whether changes in intracerebral pressure effect the electrical parameters of the cerebral cortex. We performed a s e r i e s of experiments in which intracranial pressure was elevated by injecting physiological solution into the subdural space. A rapid increase in p r e s s u r e (from 40 So 100 mm Hg) was followed by a persistent increase of resistance, probably due to the cerebral anemia that resulted. The same manipulations had almost no effect on resistance in a dead animal. In the experiments that involved a reduction of intracranial pressure by tapping spinal fluid, resistance did not change during the first 2 min, but capacitance decreased by 1%. A marked increase of resistance with a drop in capacitance that started after 8 min was caused by the aftereffects of brain dehydration, reflecting changes in the brain substance proper. Thus, we discovered from various kinds of experiments that shifts in cerebral blood flow (dilatation o r constriction of the vessels, change in intracranial pressure) by themselves effect the electrical parameters by just 0.3-0.8%. However, when the level of the metabolic processes and physiological state of the cerebral cortex change, the electrical

ELECTRICAL PARAMETERS OF THE CEREBRAL CORTEX

183

parameters are modified more rapidly. This indicates that the changes in the electrical parameters that we noted in the experiments involving action on cerebral metabolism cannot be caused by vascular shifts.

Changes in cortical impedance after electrical stimulation of the cortical surface Although electrical stimulation of the cortex has a very different effect f r o m that produced by adequate stimulations, it can be used to elucidate several aspects of cortical function (Adrian, 1936; Roitbak, 1950; Chang, 1951a, b; Burns, 1951a; and others). McIlwain (1951) showed in brain slices that electrical stimulation causes: (a) more than double the oxygen utilization; (b) increased consumption of glucose (both in the presence and in the absence of oxygen); (c) increased production of lactic acid (brain tissue in a resting organism produces very little lactic acid, but on excitation produces about 400 mmoles of lactic acid per weight p e r h; (d) consumption of high-energy phosphorus compounds with the release of inorganic phosphorus (at the start of excitation only phosphocreatine disappears; additional excitation causes the transformation of adenosinetriphosphate into diphosphate and monophosphate). Palladin (1955) found that the metabolic processes do not function in the same way in different periods of excitation. This also seems t o apply to many physicochemical changes in nervous tissue. The changes differ not only at different times after stimulation but in different functional units. Measurement of cortical impedance shows that the direction of ionic changes after electrical stimulation may undergo a biphasic change within 30-50 sec. The motor cortex was electrically stimulated for 5 s e c through implanted plate electrodes touching the surface, The animals were not anesthetized during the experiment. Square pulses with a frequency of 50 c/s were used for stimulation, the intensity of which was close to the threshold value (judging by the motor reaction, contraction of the forepaw). Resistance increased by 2-3% immediately after stimulation of supraliminal intensity w a s halted. The increased resistance lasted about 20 s e c and then gave way to lowered resistance (Fig. 87). If a second stimulation was applied soon after the f i r s t until the resistance that followed the latter ceased to increase, the shape of the resistance curve did not change. Electrical stimulation of subliminal intensity caused a slight decrease (1.5%) rather than increase in resistance (Fig. 87, C). It has been shown that cortical excitability increases not only after artificial electrical stimulation but after "spontaneous" high-amplitude discharges. According to Penfield and Jasper (1958), "epileptic" discharges generally intensify cortical excitability, resulting in unusual functional activity of the injured area. These disiharges can not be evoked by direct electrical stimulation in the absence of epileptic injury. For example, Stimulation of the temporal lobe may give rise to

IONIC PROCESSES I N THE CEREBRAL CORTEX

1a4

1

8

C

'6100 370 0

I

2

0

4

0

6 0 scc

Fig. 87. Changes in the electrical parameters of the cerebral cortex after direct A = supraliminal Stimulation thereof. R = resistance, @); C = capacitance, (IJCIF); stimulation, (f); B = two supraliminal stimulations one after the other; C = subliminal stimulation.

hallucinations if "epileptic" discharges (i.e.,in the presence of a focus of epileptic activity) appeared in this area prior to stimulation. After such discharges a trace phase s e t s in during which repeated stimulations elicit a weakened reaction. This phase was called the extinction phase by Dusser de Barenne and McCulloch (1939a). Both periods are reflected in a modification of the physicochemical properties of the cortex. There is a growth phase of resistance in the upper layers within 20 s e c that coincides with a period of increased excitability in this region of the cortex. The resistance increases during the next 20-40 sec simultaneously with the extinction period. Electrical stimulation of subliminal intensity causes only a drop in resistance. The fact that a phase of increased resistance occurs solely with supraliminaJ stimulation of the cortical surface (when a motor reaction takes place) bolsters the assumption that the increase of resistance in the area of the apical dendrites is accomp'anied by increased excitability of the pyramidal cells. The decrease in resistance with weak, subliminal stimulation may be due to the processes of excitation that a r e concentrated entirely in the superficial layers of the cortex. Direct weak stimulation excites the fibers of the superficial layer and gives r i s e to excitatory postsynaptic potentials. This local excitation of the dendrites may inhibit the cell body of the neurons located in layer IV of the cortex (Beritashvili, 1949). Intensification of stimulation may either depolarize or hyperpolarize the dendrites. In doing so resistance in the dendrite a r e a increases while the influence of the dendrites on excitability of the neuron body, in Beritov's view, decreases and, as a result, excitation is freely manifested in the deep layers of the cortex. If the intensity of stimulation is several times higher than the threshold level, the phenomenon of spreading depression of electrical activity arises. The depression period lasts 2-3 min and is accompanied by a wave of negative potential developing on the cortical surface. Cortical resistance increases 3-7% with the development of spreading depression. The point of the maximum on the resistance curve appears somewhat later than the maximum of the negative potential wave.

E F F E C T O F EXCITATION, INHIBITION AND NARCOSIS

185

According to Van Harreveld (1957), the diameter of the apical dendrites and cells increases by 15% and 5%, respectively, during spreading depression. This swelling may be caused by the migration of ions with a solvate (aqueous] membrane from the medium to the cortical cells, resulting in an increase of resistance. Thus, the resistance of the superficial layers of the cortex grows with increasing intensity of stimulation. A t first, hyperpolarization increases and the inhibitory influence of the a r e a of the apical dendrites on the deep lying cells ceases. However, with increasing intensity of stimulation, the physicochemical processes activated by it become s o vigorous that they can cause elements of the neuron to swell followed by the latter's loss of its physiological properties. THE STATE O F EXCITATION, INHIBITION, AND NARCOSIS I N RELATION TO THE ELECTRICAL PARAMETERS O F THE CEREBRAL CORTEX

Effect of stimulating drugs on the electrical parameters of the cerebral cortex Caffeine, camphor, cardiazol, and strychnine were used as stimulants in small doses that generally did not cause convulsions. The changes in electrical parameters of the brain following the administration of these substances were characterized by four phases. Fig. 88 shows the effect of intravenous injection of 20 mg/kg of body weight of caffeine. The injection w a s immediately followed by a slight decrease in resistance that soon gave way to an increase in resistance

Fig. 88, Effect of stimulating drugs on the electrical parameters of the sensorimotor cortex. C = capacitance, (WF); R = resistance, cn); (A) 20 mg/kg of caffeine; (B) 0.1 g/kg of camphor; (C) 1 g/kg of camphor; (D) "spontaneous" excitation; (E) 15 mg/kg of cardiazol; (F) 0.6 mg/kg of strychnine.

186

IONIC PROCESSES IN THE CEREBRAL CORTEX

(by 1-1.5%). This was followed, in turn, by rapid low-amplitude oscillations of the electrical parameters lasting 15-20 min. The animal exhibited greater mobility and reflex activity at this time. Approximately an hour after the injection of caffeine, resistance dropped considerably {4-6%) while capacitance rose. These four phases of change in resistance (slight drop, rise, oscillatory changes, and sharp drop) appearing after the injection of caffeine are also characteristic of the effect of other stimulants. Fig. 88, B, C, D, and E show the effects of stimulating doses of camphor injected subcutaneously at the rate of 0.1 g/kg (B) and 1 g/kg (C), of camphor injected intravenously at the r a t e of 15 mg/kg, and of a "convulsive" dose of strychnine (0.6 mg/kg injected subcutaneously). The above-mentioned four phases of changes in the electrical parameters of the cortex were more or l e s s in evidence after administration of these drugs. The first phase could not be attributed to the actual procedure of injecting the substance because the injection of physiological solution in control experiments did not give r i s e to such changes. The onset of pronounced motor excitation w a s preceded by the second phase of changes in the electrical parameters. It could be characterized as the phase of increased excitability (increase of resistance and drop in capacitance). An increase of resistance is generally a sign of impending excitation, which may occur after other agents than drugs. For example, an increase in resistance and drop in capacitance (by 1-1.5%), precedes motor excitation caused by noise in the laboratory; in fact, excitation may develop even without apparent cause (Fig. 88, D). After the injection of caffeine, the magnitude of the changes is also about 1-1.5%, after cardiazol and strychnine 2-3%. Thus, the display of excitation in an animal's behavior is preceded by definite physicochemical changes in the upper layers of the cortex that reflect the processes which accompany increased excitability. These, as already noted, consist of increased ionic mobility in the dendrite area. The third phase of changes in the electrical parameters s e t s in during the period of developing excitation (the start of oscillations). The duration of the periods of oscillations ranges from tens of seconds to seve r a l minutes. We failed to record more frequent oscillations with a low amplitude in these experiments. The figures present only the deviations with the largest amplitude. The fourth, or trace phase, is reflected in a substantial (in absolute figures) drop in resistance and increase in capacitance. Depending on the drug used, this phase appears at different times after administration and achieves different depths. It is characterized physiologically by a weakening of the exciting processes and an intensification of the inhibitory processes, physicochemically by changes in the electrical parameters of the cortex in the direction of decreased resistance and increased capacitance.

Jmpedance and electrocorticogram during excitation

ECoG's were recorded bipolarly from the same implanted plate electrodes between which the ohmic and capacitive components of impedance

E F F E C T O F EXCITATION, INHIBITION AND NARCOSIS

187

were measured. Chronic experiments were performed jointly with Dr. Smirnov on rabbits. The animals were immobilized with diplacin and given artificial respiration (through a catheter inserted into a nostril). Strychnine and phosphacol, a cholinesterase inhibitor, were used as convulsive agents. Two processes developed simultaneously in the phase preceding convulsive activity: marked increase of cortical resistance and increased frequency of oscillations in the ECoG of the cortex. In the phase of convulsive activity, resistance underwent oscillatory changes with a period of about 10-40 sec; at the same time periods of high-amplitude and highfrequency discharges in the ECoG were succeeded by periods of large slow waves combined with bursts of group discharges, The appearance of the high-frequency discharges coincided with the phase of increased resistance while periods of slow waves coincided with the phase of decreased resistance. The pattern just described was characteristic of the effect of both substances. The intravenous injection of 0.3 mg/kg of phosphacol (Fig. 89) increased resistance by 7.5% and decreased capacitance by lo%, the

Fig. 89. Impedance and bioelectrical activity of the rabbit visual cortex after the injection of phosphacol. Under each oscillogram: mark of flashing light in contralateral eye. (a) intravenous injection of 0.3 mg/kg of phosphacol; (b) intravenous injection of 10 m g / k g of atropinq; (c) second injection of 0.2 mg/kg of phosphacol. The dotted line of curve R designates the period of rhythmic oscillations of resistance (not shown). Top right, continuation of curves. Figures on the left correspond to the time designated on curve R by the figure in the circle. The -symbols on the graph are the same a s in Fig. 88.

former occurring in the form of oscillatory changes with a p e r i o d of 40-60 s e c (dotted p a r t of curve R). The ECoG also changed sharply at this time. A rhythm of 14 c/s (instead of 9 c/s before the injection of phosphacol) appeared at point 2 while at point 3 the activity was characterized by multiple discharges with a frequency of about 28 c/s. Thus, acceleration of the rhythm in the ECoG took place against a background of increased resistance. The injection of atropine (at point b) sharply

188

IONIC PROCESSES I N THE CEREBRAL CORTEX

changed both the electrical activity in the direction of evoking slow low-amplitude oscillations and the electrical parameters; resistance dropped immediately while capacitance increased. However, although the ECoG assumed its normal appearance in the following minutes (point b), resistance did not return to the original level. Fig. 90 shows another experiment with phosphacol. After the drug R

46'5

480

,

0 75

0

50

IOOsec

Fig. 90. Change in oscillations in the ECoG in time with increase and decrease in cortical resistance. (A) dynamics of resistance (R) and capacitance (C) after the injection of phosphacol (a) and atropine (b); (B) oscillatory character of the change in resistance during period 1-4 on curve A; (C) and (D) ECoG corresponding to times 2 and 3 on curves B. The symbols a r e the same as in Fig. 88.

was injected, resistance increased (by 10%). Between points 1 and 4 (Fig. 90, A) resistance increased and simultaneously underwent oscillatory changes, shown on another time scale in Fi.g. 90. The phase of increased resistance (3) is paralleled by the appearance of periodic highamplitude and high-frequency discharges in the ECoG (Fig. 90, D) while the phase of decreased resistance (2) i s paralleled by the appearance of group discharges against the background of a very slow rhythm (Fig. 90, C). Thus, the period of oscillation of resistance corresponds to the period of change in the ECoG. The results of the experiment with strychnine a r e shown in Fig. 91. Intravenous injection of the drug (arrow) increased resistance and decreased capacitance (the second arrow designates the time of another injection of strychnine; the total dose is convulsive). The second injection was followed by a marked increase in resistance and oscillations thereof with a period of 40 sec. The nature of the ECoG changes in accordance with the phase of

EFFECT O F EXCITATION, INHIBITION AND NARCOSIS

6600]4804

, u

:t

6400 470 0

189

3

60

JV IU"

lsec

,-

90 min

Fig. 91. Effect of strychnine on resistance (R) and capacitance (C) of the cortex as compared with electrical activity. = injection of strychnine (exciting dose); 3 = another injection of strychnine (convulsive dose). Bottom right on another time scale: oscillations of resistance. Top: change in oscillations in the ECoG in time with phases of decrease (Iand ) increase (II) in resistance. The symbols -are the same a s in Fig. 88.

oscillatory change in resistance. A slow rhythm combined with solitary discharges gives way to periodic high-amplitude and high-frequency discharges, which a r e in turn followed by a slow rhythm with solitary discharges, etc. The phase of resistance increase is paralleled by the appearance of a high-frequency rhythm, the phase of resistance drop by the appearance of a slow rhythm. A connection between accelerated rhythm in the ECoG and increase of resistance also is evident after other agents, e . g . , temporary asphyxia, stimulation of subcortical struckires (Fig. 92), etc. . Thus, increased cortical excitability of different origin is associated with increased cortical resistance, whereas the manifestation of excitation is associated with oscillatory changes of resistance. Two components can be detected in the change of impedance under the influence of pharmacological agents. On one hand, there is a change in the general level of impedance amounting to 8-120/0; it is determined by a shift in the metabolic "tone" of the cortex. On the other hand, there is a collateral component in the change of impedance with an amplitude of about 1-2%; it is determined by local ionic changes in the upper layers of the cortex and is intimately related to the nature of the electrical activity. Analysis of the data suggests that the oscillatory component of impedance reflects ionic changes in the a r e a of the apical dendrites caused by depolarization and hyperpolarization of the dendrites.

190

IONIC PROCESSES IN THE CEREBRAL CORTEX

cppFl I 5200 Rn 440

Fig. 92. Change in resistance (R) and capacitance (C) in upper layers of the sensorimotor cortex during and after stimulation of subcortical structures. The figures on curve R indicate the times when ECoG‘s were recorded, as shown below. The cross-hatched lines indicate the period of stimulation of the fornix. The symbols on the graph are the same as in Fig. 88.

Effect of sleep, sleepinducing and narcotic drugs on the electrical parameters of the cortex The natural falling asleep of animals in a quiet and dark place is accompanied by decreased cortical resistance and increased capacitance (Fig. 93), the changes being of the order of 2.5%. The administration of sleep inducing drugs alter the electrical parameters in the same

h

Fig. 93. Change in resistance (R) and capacitance (C) of the cerebral cortex during natural and drug-induced sleep. (A) natural sleep; (B) and (C) after the administration of sodium bromide and monobromated camphor. The symbols are the same as in Fig. 88.

E F F E C T O F EXCITATION, INHIBITION AND NARCOSIS

191

direction, but the magnitude of change is much greater (10-200/0). Fig. 93, C shows the change in electrical parameters after the subcutaneous injection of 30 mg/kg of monobromated camphor. The physiological effect of the drug is manifested in lowering of the animal's head and weakening of reflexes. This is paralleled by a pronounced drop in resistance and r i s e in capacitance. The electrical parameters of the cortex and the animal's physiological state become normal within a few hours. Sodium bromide (injected subcutaneously at the r a t e of 1 g/kg of weight) has a similar effect. This dose is soporific. Cortical r e s i s t ance decreases while capacitance increases. The changes in electrical parameters attendant on the inhibitory action of several drugs is similar to those in the trace phase of the action of stimulants. The trace phase of excitation i s apparently characterized by physicochemical changes reflecting the transition to an inhibited state. Thus, inhibition in an animal's behavior during sleep o r after the injection of sodium bromide is accompanied by a drop in cortical impedance, whereas transition to a state of excitation is accompanied by an increase of impedance. Slow "synchronous" oscillations appear in the electrocorticogram during the period of inhibited behavior. Smirnov and I made a detailed comparison of the ECoG with the resistance curve under the influence of urethane. The effect of the latter is similar to that of the abovementioned inhibitors, judging by its action both on the electrocorticogram (appearance of "synchronous" slow oscillations) and electrical resistance of the cortex (resistance drop). But there is also a difference, the mechanism of which is unclear. Capacitance does not increase as in the case of the bromides, but decreases simultaneously with resistance. Fig. 94 shows changes in the ECoG corresponding to different aspects of the change in impedance after the injection of urethane. Resistance decreases 3-5% with the onset of a narcotic condition. Capacitance changes in the same direction, but less sharply. The appearance of slow waves with a frequency of 3 osc/sec corresponds to the decrease in resistance on the electrical activity curve. It is worth noting that 15-20 min after a small dose of urethane, impedance is fully restored along with the renewal of oscillations in the ECoG at the same time that the animal emerges from narcosis. After a larger dose of the anesthetic, the slow waves in the ECoG disappear much before the impedance level r e t u r n s to normal and the physiological state is restored (Fig. 95). Thus, resistance decreases in the upper layers of the cortex upon the depression of brain activity; the slow-wave components a r e simultaneously intensified in the ECoG. The latter,phenomenon is regarded as the development of slow potentials on the neuron membranes, chiefly in the dendrite portion. The total ionic processes arising in the upper layers 'of the cortex are evidently connected .with the processes of depolarization of dendrite membranes. Depression of the central nervous system by other narcotics causes physicochemical changes in the cortex of another kind. A steady increase in resistance of 5-8% takes place during magnesium narcosis. A lethal dose produces a sharp increase of resistance of 10-15%.

192

IONIC PROCESSES IN THE CEREBRAL CORTEX

7 I

n

R

I

Fig. 94. Comparison of the changes in resistance (R) and capacitance (C) of the cerebral cortex in urethane narcosis with changes in the ECoG. Figures on Rcurve correspond to the moments of ECoG recordings. Urethane w a s administered after point 1. For explanation of symbols see Fig. 88.

Nullification of magnesium narcosis by its antagonist (calcium chloride) has the opposite effect, i.e.,it decreases resistance, the depth of narcosis being paralleled by the reduction in cortical resistance. The dynamics of increased resistance simultaneously with deepening of narcosis and the decrease in resistance after repeated injections of calcium chloride a r e shown in Fig. 96. It is noteworthy that, despite the overcoming of narcosis by repeated injections of calcium chloride, the electrical parameters do not return to the original level. With ether anesthesia, cortical resistance increases by 3-573. It is curious that some oscillations can be seen against a background of increasing resistance similar to those observed after the action of stimulating drugs. The mechanism of action of barbiturate narcosis differs, of course, from that of ether and magnesium narcosis. This difference is confirmed by the data on changes in the electrical parameters of the cortex after the action of sodium amytal (20 mg/kg). The barbiturates at first cause a slight increase in resistance in the cortex, which, as was shown, corresponds to the ordinary grow of excitability of the neuron body. This increased resistance then is followed by a slight, but prolonged decrease in resistance with increase of capacitance similar to that which occurs during light sleep. Thus, physicochemical processes of different tendency may underlie the action of different drugs which have the common property of depressing central nervous system activity. For example, ionic mobility

E F F E C T O F EXCITATION, INHIBITION AND NARCOSIS

.

35OOJ

!

10

1

193

a

30

40

min

50

5

Fig. 95. Changes with time in the restoration period of the curves of resistance (R) and capacitance (C) and in the ECoG during emergence of the animal from urethane narcosis. The symbols a r e the same as in Fig. 94. Rn

620i

610

600

590

Fig. 96. Changes in cortical resistance during magnesium narcosis and after the action of its antagonist, calcium chloride, (the arrows indicate the times of injection).

194

IONIC PROCESSES I N THE CEREBRAL CORTEX

increases in the upper layers of the cortex during natural sleep and after the administration of bromides, urethane, and barbiturates; slow waves appear simultaneously in the ECoG. Meanwhile, conditions arise that promote the realization of the modulating influence of dendrite potentials on the excitability of neuron bodies by means of extracellular currents. Other anesthetics (ether, magnesium salts) result in elevated resistance of the cerebral cortex, i.e., in the opposite ionic changes, as compared with natural sleep. This difference is apparently due to the differences in the mechanism of action of various depressants that reflect differences in the places at which the substances are applied, in the rate of exclusion of the activity of certain nerve structures, and in the depth of the effect. The increase in nervous activity is accompanied by an increase of resistance, which may be due to: (a) decrease in the number of ions in the medium caused by their migration into the intracellular space; (b) greater concentration of ions on polarizable boundaries; (c) increase in resistance of the cell membranes; (b) and (c) may be connected with the processes of hyperpolarization of the dendrite membrane.

Changes in cortical resistance compared with phases of nervous activity The dynamics of nervous processes in the cerebral cortex may be investigated by analyzing the convulsive states caused by over stimulation of a receptor. The specific reaction of a rat to sound (Krushinskii's model) can be used as such a model. One or two waves of strong motor excitation, ending in a seizure with clonic and tonic convulsions, appear in the rat in response to the inclusion of a loud sound for several seconds. The rat's response to sound has several phases. The animal remains outwardly calm for a few seconds after the bell sounds; this is the latent period of the reaction. In 5-10 s e c the rat suddenly exhibits strong motor excitation, which lasts 8-15 sec. A seizure (the so-called single-wave reaction) may set in after this initial wave. In other cases, the first wave of excitation is followed by temporary calmness for 5-10 sec, despite the fact that the bell is ringing. A second wave of excitation lasting 7-25 sec suddenly appears which either damps off o r develops into a deep seizure. A long period of restoration (20-30 min) initiated by catatonia follows termination of the seizure. This model of an epileptoid attack m&es it possible to observe a complex succession of states of animal excitation and inhibition in a short interval of time and to differentiate the phase changes quite precisely. The ohmic and capacitive components of impedance of the rat motor cortex were measured while the bell was sounding, during an attack, and in the period of restoration of the animal's functional state. A total of 67 experiments on 18 rats were performed*. *Mr.Z. Apanasenko assisted in this series of experiments.

EFFECT O F EXCITATION, INHIBITION AND NARCOSIS

195

Implanted electrodes (silver plates 1.5 mm in diameter) were placed on the motor region of a hemisphere (space between electrodes 1.7 mm). The dura mater under the electrodes w a s preserved. The experiment was performed 8 days after the operation. The animal was placed in a spacious box with a bell in an adjoining compartment. The animal w a s able to jump and move about freely. Rubber balloons in the double bottom of the box were connected to a Murray capsule to help in recording a mechanogram of the movements. A flexible wire coiled into a spiral led from the electrodes to the measuring apparatus. The electrical parameters of the cortex were measured at low (10,000 c/s) and high (30,000 c/s) frequencies. The background changes in impedance were measured every 5-10 min for an hour prior to starting an experiment with the bell. Lissajous figures on the screen of a cathode oscillograph were photographed on film (8 frames per sec) while the bell was ringing and the rat w a s in a state of motor excitation. Fig. 97 presents an experiments with a "single-wave" rat; the

2 .4

6

8

10

12

14

16

18

sec

Fig. 97. Resistance (R) and capacitance (C) of the r a t cortex in different phases of nervous activity (after stimulation with sound). (a) latent period; (b) motor excitation; (c) seizure. At the zero time sounding of a bell. The symbols a r e the Same a s in Fig. 83.

electrical parameters were measured at a frequency of 10,000 c/s. A seizure occurred after the initial motor excitation. Resistance began to increase during the latent period, when the bell had already sounded but the animal w a s outwardly calm. This apparently reflected its increased excitability. When resistance reached a certain level, the r a t suddenly exhibited strong motor excitation. Resistance continued to increase at this time and oscillatory changes appeared therein as capac itance decreased. However, impedance dropped sharply 2-3 s e c before a seizure, which developed with low resistance. In the experiments described in the preceding paragraphs, the drop in impedance was usually associated with the development of inhibition.. The drop in impedance as the excitation phase turned into the seizure

196

IONIC PROCESSES IN THE CEREBRAL CORTEX

phase suggests that a seizure may occur at a time when the cortex is functionally inhibited (according to published reports, a seizure is caused by the passing of excitation into the subcortical structures). It is worth noting that changes in the electrical parameters preceded the physiological manifestation of nervous activity by 2 - 3 sec. The time when the various phases of nervous activity began differed somewhat from experiment to experiment, with corresponding phase changes in the electrical parameters. For example, if resistance began to increase from the very first second after inclusion of the bell, the latent period would be short (3-5 sec) and the phase of motor excitation would quickly s e t in. On the other hand, if resistance did not begin to increase until 3-4 sec after inclusion of the bell, the latent period would be extended to 7-10 sec. Thus, a change in cortical resistance during the latent period makes it possible to predict just when the animal will have a motor reaction. A sudden drop in resistance toward the end of the motor phase is invariably a sign that a seizure will occur within 2 - 3 sec. The same pattern of impedance changes was revealed in the "two-wave reaction'' of an animal. Resistance increased by 0.5-1.5% during the latent period of the physiological reaction and continued to increase in the first phase of motor excitation. Toward the end of this phase, a sharp change set in toward a decrease in resistance. This forecast the onset of calmness in the animal's behavior, during which time resistance r e mained 0.5-1% below the original level. A sign of impending termination of the period of calmness is a new increase in resistance, which indicates the s t a r t of the second phase of motor excitation. Resistance decreases sharply by 3-5% before a seizure. In designing these experiments, we were naturally concerned that a change in impedance might be due to the rat's movements and expenditure of energy on mechanical work (jumping). We therefore ran a control s e r i e s of experiments. The animal was exposed to sound stimulation while its body w a s tightly secured to a bench. The phases of the reaction were judged from the movements of one paw. This series of experiments revealed the same pattern of changes in the electrical parameters. This pattern of changes in impedance was observed whether a lowfrequency (10,000 c/s) or a high-frequency current (300,000 c/s) was used. Resistance in the different phases of the reaction changed the same way with either frequency. This suggests that in this case ionic changes in the cortical substrate a r e largely due to change in the total number of mobile ions and not to change in the conditions of polarization on the boundary surfaces. The results described above were obtained in the overwhelming majority of the experiments (24 out of 27). After the onset of a seizure, the electrical parameters of the cortex continued to undergo complex trace changes, which were of greater magnitude than while the bell w a s sounding. The direction of these changes was related to physiological manifestations. Fig. 98 represents the trace phases occurring after a seizure (the time scale is in minutes, hence the figure shows only trace changes, not changes that occurred as the bell was sounding). Trace changes can be arbitrarily divided into four phases: I = emergence from an attack; I1 = development of catatonia;

EFFECT O F EXCITATION, INHIBITION AND NARCOSIS Ac, %

At% 6 -

197

b

5 -

4 3 2 I -

0% t -

R

z 3 U -

Fig. 98. Trace changes in resistance (R) and capacitance (C) of the r a t cerebral cortex after sound stimulation (excitation phase not shown). (a) before an experiment; (b) emergence from an attack; (c) period of catatonia; (d) weakening of catatonia.

I11 = termination of catatonia; IV = general inhibition. T h e r e was an increase in resistance (about 25%) after the s e i z u r e w a s over; the s e i z u r e , as noted above, occurred against a background of d e c r e a s e resistance. The phase of increased cortical resistance gave way to a phase of sharply decreased resistance (about 20%), at which time catatonia set in. An increase in r e s i s t a n c e in phase I11 p r e ceded the disappearance of catatonia. In phase TV r e s i s t a n c e decreased, reflecting the generally inhibited behavior of the animal. A comparison of these changes in the electrical p a r a m e t e r s of the cortex with the nature of the ECoG, based on Vassilyeva's data (1955, 1958) obtained from rats under s i m i l a r experimental conditions, suggests s o m e parallels. During the latent period of the reaction, cortical r e s i s t a n c e i n c r e a s e s , low-amplitude oscillations are accelerated, excitability rises. During the period of motor excitation of the animal, cortical r e s i s t a n c e continues to increase while the frequency and a m plitude of the oscillations in the ECoG increases. However, t h e r e is s o m e divergence between the ECoG data and the dynamics of cortical r e s i s t a n c e upon the onset of a seizure. Cortical resistance begins to increase just before a s e i z u r e while high-frequency activity r e m a i n s predominant in the ECoG. Slow waves appear in the ECoG after a s e i zure, reflecting decreased cortical resistance. It was not our purpose to elucidate cortical-subcortical relations during the convulsive reactions of rats. We have described only the

198

IONIC PROCESSES IN THE CEREBRAL CORTEX

dynamics of the ionic changes in the cortex during different phases of nervous activity, but it i s safe to assume that the underlying portions of the brain play a part in these changes. The dynamics of cortical resistance during an experimental epileptoid attack may reflect the migration of f r e e ions between the cells and extracellular medium as a result of the development of the corresponding nervous processes in the cells. However, one cannot exclude the role played by purely metabolic mechanisms, which can change the ionic composition of the medium by the formation of metabolic products. Just as in the case of pharmacologic actions, two components can be distinguished in impedance changes. During the initial phases the change i s from 1-3% and is very temporary. It is caused by ionic changes in the upper layers of the cortex. Change in the general level of impedance (8-15%) i s dominant in the trace phases and represents changes in the metabolic "tone" of the cortex. Of great interest is the fact that the ionic changes in the cortex occur 2-3 sec before any physiological manifestation of the reaction. The inflow of rapid afferent impulses from the sound receptor evokes ionic changes in the motor cortex. They a r e initially manifested in increased excitability of the neuron bodies, which prepares the latter for greater activity (phase of motor restlessness). The ionic processes in the dendrite a r e a change direction sharply 2-3 s e c before a seizure; ionic mobility increases. An impetus to this reversal is the appearance of new active hormonal substances which influence dendrite activity. All this may result in a lowering of cortical neuron activity with release of the subcor t ical mechanisms of convulsive seizure

.

DYNAMICS O F IONIC CHANGES IN THE CEREBRAL GANGLION OF THE CRAB

Investigations of cortical impedance in warm-blooded animals have shown that ionic changes in the area of the apical dendrites may have a different tendency and anticipate change in the physiological state. It has been conjectured that ionic changes may a r i s e specifically as a result of the appearance of active substances following stimulation of the neuroglia. The ganglia of arthropods a r e quite rich in hormones, which may be stored in glial cells and released during excitation. According to Enami (1951a, b), stimulation of the crab's afferent structures intensifies secretion in the ganglia. Electrical activity of the crab's ganglia is characterized by the predominance of high frequencies over low. Koshtoyants et a l . (1954) showed that a chemical medium created during the metabolic process may be the source of constant stimulation of the ganglion of the silkworm cocoon. With this in mind, we studied the dynamics of ionic changes in the cerebral ganglion of the crab. Electrodes to measure impedance were placed in such a way that the force lines of current flowed chiefly through the central portion of the cerebral ganglion, which has a considerable concentration of nerve fibers and dendrites. The electrodes consisted of silver plates 0.8 mm in diameter. They

DYNAMICS O F IONIC CHANGES

199

were brought into d i r e c t contact with the ganglion on two opposite s i d e s through trepanations in the shell. Ganglion resistance and capacitance w e r e measured with a current at 10,000 c/s. The following experiments were performed: (a) stimulation of the c r a b ' s eye with a flashing light (the beam was narrowly focused on the eye; special efforts were made to prevent the ganglion region from becoming heated); (b) stimulation of the olfactory receptor (antenna); (c) application of cocaine and acetylcholine to the ganglion, use of chloroform anesthesia. Experiments were performed with lymph circulation both preserved and excluded (by removing the heart). The animal's state of excitation o r inhibition was judged from the motor activity of the extremities, whose connection with the central nervous system was left intact. The motor activity of the c r a b extremities is determinded by the function of the subesophageal ganglion, which is influenced by the c e r e b r a l ganglion. The latter inhibits the f o r m e r . The motor activity of a c a r i d s increases after removal of the c e r e b r a l ganglion (Bethe, 1897). In this ganglion are concentrated the dendrite branches, whereas the neuron bodies responsible for motor function are in the subesophageal ganglion. This should be kept in mind when comparing the condition of the animal's c e r e b r a l ganglion and motor activity .

Silwzmntioiz of iotzic changes in the ganglion aftev veceptov stirnulcction The electrical p a r a m e t e r s of the c r a b c e r e b r a l ganglion undergo slow oscillatory changes, with a period of about 2-3 min, and occasionally m o r e rapid changes, with a period of about 35 s e c . Stimulation of the antenna caused a 1-8% d e c r e a s e in resistance during the f i r s t 20-30 s e c and a proportional d e c r e a s e in capacitance; subsequently, however, the maximum d e c r e a s e in capacitance occurred 10-20 s e c after resistance had already begun to change in the opposite direction. Resistance then increased for 5-15 min as rhythmic oscillations sometimes appeared with a period of about 2 min. T h r e e experiments are illustrated in Fig. 99. If the specimen was stimulated repeatedly o r excitability was poor, the absolute magnitude of the changes decreased and restoration of the initial level required m o r e time. Thus, r e s i s t a n c e decreased even while the afferent pathways were being stimulated. When stimulation was halted, it took a long time bef o r e the electrical p a r a m e t e r s , frequently oscillatory in nature, w e r e r e s t o r e d . It seemed worthwhile to u s to verify the phenomenon with adequate stimulation of another receptor v i z . the eye. We found that intermitte'nf illumination of the eye (every 5 s e c ) caused ganglion r e s i s t artce to drop as long as the stimulation continued; meanwhile capacitance increased. When stimulation was halted, the electrical p a r a m e t e r s returned to the original level, the time required being directly proportional to the duration of illumination. The dynamics of the p r o c e s s is illustrated by the experiment shown in Fig. 100. Illumination of the eye for 3 min decreased ganglion r e s i s t ance by 1.1%while capacitance increased by 0.9%. When illumination

200

IONIC PROCESSES IN THE CEREBRAL CORTEX

230

I

. 1 , , , , , , , , , , , , , 0

I

2 8

4 5

8

7

6

9 10 N h? N m i n

Fig. 99. Changes in resistance (R) of the crab cerebral ganglion after stimulation of the antennae. 1, 2, 3 = different observations.

Fig. 100. Effect of illuminating the crab eye on electrical resistance of the cerebral ganglion. The arrows designate the times when each illumination was begun. The duration of stimulation is indicated by the figures below the cross-hatching.

was halted, resistance returned to.the original level within 2 min. Brief illumination for 7 s e c decreased resistance by 0.2% without any effect on capacitance; subsequent restoration required 20 sec. When the interval between the two series of stimulations was insufficient for restoration of the electrical parameters to the original level, a second s e r i e s caused the ordinary decrease in resistance, i . e . , the effects of the stimulations were summed. In some experiments it was poss.ible to observe a quantitative relationship between the duration of light stimulation and magnitude of the change in resistance. For example, 1 min of illumination decreased ganglion resistance by 1.7R;2 min of illumination decreased it by 2 . 3 Q 3 min by 3 n . Thus, there was a summation of the ionic changes in the ganglion both when the time of illumination was lengthened and when the stimulation was repeated. Fatigue of a specimen is shown by lengthening of the latent period of the reaction to the light. For example, a change in the electrical parameters of a specimen with good excitability became noticeable 3-5 sec

DYNAMICS OF IONIC CHANGES

201

after the start of illumination; when fatigued, the latent period of the specimen was lengthened to 12 sec. Another indication of fatigue is disappearance of the summation phenomenon. For example, in a fresh specimen exposed to 3 min of illurnination, the change in resistance (decrease) increased throughout this period of time. In a fatigued specimen, resistance decreased for 35 s e c and then began to increase, despite continuing stimulation of the receptor. Illumination had no effect on the electrical parameters of a specimen with very poor excitability. Decreased ganglion resistance indicates an increase in the number of free ions p e r unit of ganglion volume. The flow of afferent impulses apparently changes the ionic equilibrium between the medium and the cell which was present in a resting state, increasing the outflow of ions into the extracellular space. It is interesting to compare this with the observations of P r o s s e r (1934a, b), who recorded the frequency of impulses in the ganglion of the crayfish abdominal nerve cord following stimulation of the photoreceptor. With increase in time of illumination from 0 to 4 sec, the frequency of impulses r o s e from 70 to 120 c/s, .i.e., the longer the stimulation, the higher the frequency of impulses. This fact plus the phenomenon of prolonged summation of ionic changes supports the assumption that some active substances are liberated in response to the inflow of afferent impulses, the longer the stimulation, the greater the amounts liberated..

Effect of application of potassium chloride, cocaine, acetylcholine solutions and anesthesia on electrical parameters of the ganglion The application of tiny drops of a solution of potassium chloride in a concentration of 580 mM decreases ganglion resistance by 10-20% while increasing capacitance. This effect lasts 4-10 min and is then followed by restoration of resistance to the .original level. Application of the same amount of s e a water reduces ganglion resistance by 1%.Washing the ganglion r e s t o r e s its electrical properties. Thus, an excess of potassium ions in the external solution causes a significant decrease in ganglion resistance. This appears to be due both to depolarization of the neurons induced by the potassium chloride and to the appearance of a larger amount of these ions in the ganglion substrate. Solutions of cocaine (2% in sea water) and acetylcholine ( l o r 7 ) have a central exciting effect, which is manifested in intensified motor activity of the crab's extremities and in increased tone of the muscles of the extremities. Illumination of the eye (with a light of ordinary intensity) may also stimulate motor activity. This never occurs in a crab not exposed to cocaine. The application of cocaine causes a sharp increase in ganglion resistance of 10-15% (and a drop in capacitance). The appearance of oscillatory changes with a period of about 20-30 sec against a background of increased resistance is quite characteristic (Fig. 101). The phase of increased resistance is featured by intense motor activity of the extremities and an increase in their tone. A similar effect is observed after the application of acetylcholine (Fig. 102). Following a temporary drop in resistance, due, in our opinion,

202

IONIC PROCESSES IN THE CEREBRAL CORTEX

Fig. 101. Oscillatory changes in the resistance (R) of the crab ganglion after the application of cocaine.

Fig. 102. Effect of acetylcholine on resistance (R) and capacitance (C) of the crab cerebral ganglion. Acetylcholine (loe3) applied the 2nd min; motor excitation noted between the 3rd and 5th min; depression of activity from the 5th min.

to the shunting action of the drops,, resistance increases about 10%. If the drops are very carefully applied a little to the side of the electrodes, there is no initial drop in resistance. The extremities move about vigorously during the period of intensified resistance. There a r e also oscillatory changes in resistance, the frequency of the oscillations being even higher than after the application of cocaine (22 osc/min). That is to say, the period of oscillations lasts approximately 3 sec. (Sometimes there a r e respiratory pulsations of the crab ganglion, but they are more frequent than the oscillations of resistance, for the crab breathes at the r a t e of 120 per min.) A sharp change, in the direction of a decrease, may occur on the resistance curve after 2-3 min (Fig. 102). The extremities relax at this time, i.e., signs of central inhibition appear. In other cases, h e excitation phase lasts 8-12 min and is accompanied by fibrillation of the extremities, during which time r e s i s t ance increases steadily. Fibrillation ceases when resistance begins to drop. We recorded the electrical activity of the cerebral ganglion from the same electrodes in the phase of increased resistance after the

CONCLUSIONS

203

application of acetylcholine. The nature of this activity varied with the period equal to the period of resistance oscillations. Slower waves in the electrogram were followed by high-frequency activity paralleling the transition from decreased to increased resistance. This finding was very similar to the data obtained from the rabbit cortex after exposure to the cholinesterase inhibitor phosphacol, which promotes the storage of acetylcholine . Experiments with chloroform showed that ganglion resistance increases during the first phase, which is paralleled by motor excitation. The subsequent course of the changes in ganglion resistance may be either of the following: (1) Resistance decreases against.a background of relaxation of the extremities and absence of a reaction to stimulation of the antennae, with resistance restored when the animals come out of the anesthesia; (2) A large dose of chloroform results in a steady increase in r e s i s t ance. Paralleled at f i r s t by motor excitation, resistance continues to increase, but eventually the crab ceases to move. This experiment usually ended in the animal's death, implying that a lethal dose of the anesthetic causes ionic changes in the same direction as in excitation. Thus, increased resistance in the cerebral ganglion (chiefly in the dendrite area) is associated with the effect of substances that excite motor activity in animals (cocaine, acetylcholine). The reasons for such change in resistance a r e the metabolic alterations caused by the application of these substances. Increased resistance in the dendrite a r e a may well signify termination of the inhibitory action of the c e r ebral ganglion on the subesophageal ganglion and excitation of the latter, as reflected in the animal's motor activity. It is interesting to compare these results with the data on the brain of a warm-blooded animal. The manifestation of motor excitation is preceded by an increase in resistance of the upper layers of the cortex. In other words, in both cases increased resistance in the apical dendrite a r e a signifies termination of their inhibitory action on the function of the neuron bodies in the motor regions. Decreased resistance in the dendrite a r e a is preceded by inhibition of motor activity which, in the case of the crab, means intensification of the inhibitory action of the cerebral ganglion on the subesophageal ganglion. CONCLUSIONS

The dynamics of ionic processes in tissue is reflected in changes in its conducting properties. The complexity of cortical structure makes it difficult to evaluate the measurements of its conductivity. The substrate of the gray substance is polyphase ,with respect to the electrical properties of the system. The cell elements, in particular, are surrounded by a membrane with high resistance. Therefore, the field of the current conducted through the gray substance from an external source passes mainly through the interstitial fluid.

204

IONIC PROCESSES I N THE CEREBRAL CORTEX

The following factors attribute to changes in the impedance of the gray substance: (1) change in properties of the interstitial fluid; (2) change in resistance (membrane) of cell elements; (3) change in the volumes of the elements that make up the gray substance; (4) shifting of ions between the extracellular and intracellular space; (5) change in the concentration of ions on polarizable boundaries. These changes a r e reflected in the alteration of both the ohmic and the capacitive components of impedance. Resistance and capacitance change, as a rule, in opposite directions, which accounts for the relationship between the absorption charge (which creates the capacitance) on the boundary phases and the total number of ions in the solution. In vivo observation of the dynamics of ionic changes in the cortex is effected by measuring the resistance and capacitance in the tissue between implanted electrodes by means of an alternating current with a frequency of 10,000 c/s. This method makes it possible to record these parameters simultaneously during a rapidly changing process. The ionic composition of the gray substance of the hemisphere, investigated in wakeful animals, undergoes slow fluctuations, which cor relate with fluctuations of cortical excitability (judging from reactivity to rhythmic receptor stimulation). Significant ionic changes occur after a change in oxygen and carbohydrate metabolism. A control s e r i e s of experiments involving changes in blood pressure and intracranial pressure showed that the hemodynamics of the brain as a physical factor is reflected in an alteration of cortical impedance of about 0.5%, whereas the ionic changes associated with alteration of brain metabolism a r e reflected in a 5-15% change in the gray substance. Resistance measured with tangential arrangement of the electrodes on the cortical surface and with the electrodes located close together is determined chiefly by the ionic processes in the upper layers of the cortex . An investigation of changes in conductivity of the dendrite layer in different functional states of the cortex is of value in analyzing the mechanism of dendrite influence on the excitability of neuron bodies. Excitation of the apical dendrites is followed by a decrease in resistance of the upper layer of the cortex while capacitance increases. This may be due to the migration of ions into the extracellular fluid caused by mass depolarization of the dendrites. Excitation of the deeper lying neurons is paralleled by the increase of impedance in the upper layer, i.e., by a decrease in the number of f r e e ions in the dendrite area. The r i s e in resistance in the dendrite a r e a after excitation of the deep lying neuron bodies throws light on the role of the dendrites in maintaining the intracellular ionic composition by capturing from the medium the ions needed to execute neuron function. However, increased resistance in the a r e a of the sensorimotor cortical dendrites is preceded by the development of motor excitation, whereas decreased resistance of the dendrite a r e a occurs several seconds before the level of behavioral activity is lowered. This was clearly shown by the experiments in which rats were stimulated with loud sounds. During the first few seconds of stimulation, resistance in the upper layers of the cortex began to r i s e , even though the rats had not as yet exhibited any motor activity. This

CONCLUSIONS

205

means that the physicochemical changes in the apical dendrite a r e a take place before the cell bodies of the pyramidal cells a r e excited. The afferent impulses reaching the dendrites via specific and nonspecific pathways presumably activate there the ionic processes which a r e reflected in changes in impedance. The experiments on the crab cerebral ganglion show that ionic changes in the axodendrite p a r t of the ganglion a r e summed after prolonged stimulation of the afferent pathways and that summation is impaired when the specimen becomes fatigued. In this case, the ionic changes in the ganglion may be caused by the neurosecretory activity of glia, a characteristic of arthropods. The magnitude of cortical impedance changes in relation to the functional state of the animal. During a period of increased excitability of the central nervous system, electrical resistance of the cortex inc r e a s e s while high-frequency oscillations a r e intensified in the ECoG. Resistance continues to increase during the period of excitation and assumes an oscillatory character (with a period of 10-40 sec). Oscillations of different appearance a r i s e in the ECoG at this time. Highfrequency waves are followed by slower waves. The former originate in the phase of increased resistance, the latter in the phase of decreased resistance. When the animal's level of activity is lowered, resistance decreases in the upper layers of the cortex while slow waves appear in the ECoG. Two elements a r e to be distinguished in the changes in cortical impedance. One marks the general level of impedance and changes more or less simultaneously with the animal's physiological state. The other element reflects the local ionic processes and changes within much narrower limits, and it is intimately connected with changes in the electrical activity of the cortex. Changes in the appearance of the oscillations in the ECoG a r e due to local changes and not to the general level of impedance. In other words, the appearance of higher frequency oscillations in the ECoG is always associated with a phase of increased resistance regardless of the absolute level from which the increase begins. Impedance in such instances reflects the local ionic changes connected with the activity of the nerve elements and with the dynamics of ion movement between the extracellular and the intracellular spaces. The appearance of slow potentials in the ECoG, which, according to modern views, reflect local postsynaptic potentials, i s paralleled by a decrease in resistance of the dendrite layer, which causes an increase in the extracellular current between the apical dendrites and the neuron body and influences neuron excitability. Thus, the oscillatory element of impedance reflects the local ionic processes created by the excitation of certain structures and by the inhibition of others. The other element of cortical impedance reflects changes in the total number of ions, which is determined by the metabolic "tone'' of the cortex. For example, the inhibition of tissue cholinesterase causes a marked increase in cortical resistance, which reflects trophic changes in brain metabolism. Cholinesterase is also present in glial cells whose metabolism affects the magnitude of resistance. Excitation of the apical dendrites results, therefore, in increased

206

IONIC PROCESSES IN THE C ER EB R A L CORTEX

ionic mobility in the upper l a y e r s of the cortex. Conversely, excitation of the deep lying neuron bodies is preceded by decreased ionic mobility in the apical dendrite area. Total ionic changes occur in the cortex following a change in its metabolic "tone".

20 7 CHAPTER IX

E F F E C T O F IONIZING RADIATION ON THE INFRASLOW POTENTIAL OSCILLATIONS AND ELECTRICAL PARAMETERS O F THE BRAIN

The sensitivity of tissue to ionizing radiation is generally g r e a t e r when it i s already highly sensitive to chemical factors in the environment. The s t r u c t u r e s characterized by an auto-oscillatory, infraslow p r o c e s s have high chemical sensitivity as well as radiosensitivity. This principle i s particularly applicable to muscle tissue; X-irradiation causes g r e a t e r changes in smooth muscles than it does in skeletal muscles (Aladjalova, 1960a, b). It holds t r u e for changes in both functional and s t r u c t u r a l properties (judging by the electrical p a r a m e t e r s ) . The chemical sensitivity of muscle tissue is closely connected with the nature of its macromolecular structure. It i n c r e a s e s after denervation of muscle and its structural properties change at the s a m e time (Aladjalova, 1950a). The radiosensitivity of denervated muscle increases with the increase of its chemical sensitivity. If the indicator of radiosensitivity of isolated muscle is the critical dose the irradiation of which causes i r r e v e r s i b l e changes in excitability, the dose for smooth muscle ranges from 2230,000 R, for skeletal muscle it ranges f r o m 40-50,000 R. The critical dose f o r denervated skeletal muscle drops to 30,000 R. If other agents (chemical, temperature) are combined with radiation, the critical dose is even s m a l l e r , i . e . , the destructive nature of the changes is revealed sooner. Therefore, the radiosensitivity of isolated muscle v a r i e s with its chemical sensitivity, which, in turn, is related to its macromolecular (high-polymer structure), and with the effect of additional agents on the muscle. This last circumstance is of particular significance when considering the effect of radiation on the whole organism of a warmblooded animal in which an inevitable chain of central reactions brings about chemical changes in the internal medium and temperature changes. Analysis of the effect of radiation on b r a i n structure is complicated by the need to take a host of phenomena into consideration. The role of the nervous s y s t e m is determined both by the p r i m a r y injury thereto and by the reaction t o pathological changes in other organs. Thus, var i o u s methods of irradiation must be utilized: whole-body irradiation, local irradiation of peripheral organs, irradiation of the trunk with the head screened, local irradiation of the head, and narrowly local irradiation of the individual brain structures. The mass of experimental material that we obtained by X-irradiating animals with different doses, and at different sites, which h a s been fully described in a s e p a r a t e paper (Aladjalova, 1960b), indicates that the principle underlying the relationship between chemical and radiosensitivity apparently also holds t r u e for neurons of the central nervous system. Moreover, it may also apply t o individual elements of the neuron. The dendrites, being m o r e chemically sensitive than the axons,

208

E F F E C T OF IONIZING RADIATION

may likewise be more radiosensitive. This assumption i s supported by data revealing the effect of radiation on impedance of the upper layers of the cortex and on infraslow potentials in different brain structures. The parameters of the ISPO in the cortex and in the hypothalamus change regularly in relation to the dose and irradiation site. These parameters a r e affected both by direct and by remote X-irradiation. Direct exposure (of the head) results in depression of the ISPO in the cortex, the duration increasing with the size of the dose. The initial depression resulting from doses of 300-1000 R (rate 32 R/min) increases the amplitude of the IsPo within a few hours; their frequency increases between the 4th and 20th day from 8 to 12 osc/min. These phases do not occur simultaneously in different regions of the cortex. Remote irradiation (of the abdominal region with doses of 300-1000 R (rate 32 R/min) increases the amplitude and regularity within the first 24 h and then (up to the 7th day) causes the ISPO in the sensorimotor cortex to accelerate from 8 to 14 osc/min. The ISPO's a r e intensified in the cortex and hypothalamus at the same time. This i s preceded by a period of intensified electrical activity in the hypothalamus. The phase of intensified electrical phenomena in the hypothalamus s e t s in early, also after whole-body irradiation (1000 R). The investigations of Livanov (1956-1959), who used the method of conditioned reflexes and electrophysiology, provide evidence of increased excitability of the hypothalamus after whole-body irradiation. There is also information available on changes in hypothalamic function after whole-body irradiation even with a small dose of 50 R (Lebedinsky et al., 1959). Prolonged excitation of the hypothalamic a r e a may result from the inflow of afferent impulses (chiefly from the interoceptors) or from direct irradiation with small doses. Livanov (1959) showed that increased excitability of subcortical structures appears even when inhibition has already developed in the cortex, i.e., by the end of the f i r s t day after whole-body irradiation. Severe impairment of the regulatory activity of the brain stem (like that following hyperexcitation) is regarded as one of the possible causes of radiation death. The mechanism of intensification of the ISPO in the cortex i s intimately associated with the hypothalamic system which activates the hormonalmetabolic function of the brain. The sensitivity of this indicator to radiation largely reflects the influence of the latter on the hypothalamus. Irradiation of different sites is followed by intensification of the ISPO not only in the cortex and hypothalamus, but also in the medial and intralaminar nuclei of the thalamus, which give r i s e to nonspecific thalamocortical projections, and in some other nuclei of the thalamus ( e . g . , in nucl. lateralis posterior). Infraslow rhythms do not a r i s e in the specific nuclei of the thalamus (lateral anterior and lateral ventral). They a r e also found in the central gray substance. Thus, soon after irradiation a period of intensified electrical phenomena in the hypothalamus regularly occurs during which the hormonal activity of the hypothalamus is stimulated, as manifested in the intensification or appearance of ISPO in many p a r t s of the brain. ISPO's are also intensified in the brain and hypothalamus after "stress"

DIRECT AND REMOTE RADIATION

209

reactions caused by nonradiation factors. The question a r i s e s of whether the increased frequency of the ISPO after irradiation is attributable to the development of a nonspecific "stress" reaction. In the case of irradiation, e . g . , of the abdomen of an animal with small doses, it is quite probable that the regulatory and neuroendocrine functions of the brain are included by reflex action (until the appearance of the toxic phase) as in the ordinary "stress" reaction. Sublethal doses of irradiation actually stimulate the production of certain hormones in the body (Betz, 1956; Moiseyev, 1957) that a r e characteristic of the "stress" reaction. However, it is characteristic of radiation injury that this regular reaction has neither compensatory nor protective significance as in the case of exposure to a nonradiation agent, and it can only aggravate the injurious effects of the radiation. The possibility of such an effect aggravating radiation injury is confirmed by experiments in which the postirradiation injection of hormones - testosterone and cortisone (Ellinger, 1947) and the adrenocorticotropic hormone (Betz, 1956) - aggravated the course of radiation injury. A feature of radiation injury i s that active chemical agents (hormones) in the body may help to shorten the "latent" period of radiation injury. In connection with the undoubtedly important role played by the hypothalamic area in the reaction of the nervous system to irradiation, it is interesting to analyze the effects of direct radiation on the hypothalamus. Highly local irradiation of a small portion of nervous tissue with a microemitter is a method that makes analysis of the direct reaction of the substrate to irradiation possible. Liberman (1958a) described a method of irradiation through a microelectrode into which a @-emitterhas been inserted. The isotope is applied on the thin end (5-10 p) of a glass fiber and it is coated with a layer of lacquer 5-10 CI thick. With the specific activity of the 8-emitter equal to 1 pC/cm", the r a t e of absorbable radiation i s of the order of 10 rad/min.* We used microirradiation of the preoptic nucleus of the hypothalamus in which there a r e active processes of neurosecretion. A glass cannula was inserted into the preoptic nucleus of rabbit brain with a stereotaxic instrument. Ten days after the operation, a thin-walled capillary 100 p in diameter was inserted into this cannula. The capillary, which contained a phosphor isotope, was covered with a thin layer of lacquer. To avoid injuring nervous tissue, the cannula and capillary were so designed that the tip of the emitter remained 50 p before the level of the cannula tip. The activity of the emitter was determined by placing it in a cannula similar to the one inserted. The activity came to 3.5.1OW2pC. The absorbed dose rate (M) of p-radiation at distance R from the point source can be expressed as follows (Liberman, 1958b):

where A is the activity of the source, L is the coefficient, varying B with the energy of isotope radiation (for 32P,L = 4.1*10+), p(R) is the B *The experiments were performed jointly with Dr. Koshtoyants.

210

E F F E C T O F IONIZING RADIATION

DIRECT AND REMOTE RADIATION

211

function describing the weakening of the absorbed dose of @-radiation in relation to the distance from the point source; Q is a constant depending on the properties of the medium. Since the accuracy of calculation required can scarcely exceed lo%, Loewinger's simplified formula can be used: 4R Ae"cLR p(R) = L; then M = ___ R R '4-n; L8J

112

where CI is the coefficient which for soft tissue i s taken as 9 cmz/g; e is the base of natural logarithms. The power of our emitter was 0.02 rad/min o r 20 mrad/min at a distance of 0.1 mm from the emitter. It was calculated from Liberman's formula (1958a) with the following substitutions: A = 3.5.10"" pC; p = 9 cm2/g; R = 0.1 mm; L = 4.1.10"4; p = 1 mm. This dose is far in excess P of the radioactive background of the organism, which amounts to about 20 mrad/year. Electrical activity of the sensorimotor cortex was investigated from the time the emitter was inserted into the preoptic nucleus, where it remained 3 days. The absmbed dose w a s 30 rad/day for 24 h. It w a s inadvisable to keep the emitter there any longer because by the fourth or fifth day w e could see some of the isotope s t a r t to penetrate into the body, possibly due to disintegration of the thin protective coat of lacquer. Experiments with local irradiation of the preoptic nuclei were p e r formed on 4 rabbits. Within a few minutes after insertion of the emitter, when the dose came to 0.3 rad, the ECoG showed bursts of high-amplitude oscillations typical of the effect resulting from stimulation of subcortical structures. N o effect was evident in the contralateral hemisphere. This periodic activity in the form of bursts w a s noted in two animals throughout the f i r s t day (Fig. 103). It disappeared the second day, and on the third day only slow oscillations with a frequency of 5 c / s could be seen in the ECoG. The ISPO's were intensified 3-5 h after insertion of the 8-emitter into the sensorimotor cortex of both hemispheres and the frequency of the rhythm r o s e to 1 2 osc/min, hi two other cases, ISPO of the same high frequency appeared on the second day after irradiation and remained 6 days after the emitter w a s removed; they were more pronounced in the ipsilateral hemisphere, This effect has been caused by excitation of the neurohormonal function of the hypothalamus, which seems to be possible either as a result of reflex reaction to stimulation of receptor fields or by the direct effect of radiation on the preoptic nucleus. The results suggest that the direct effect of low doses (0.3-30 rad) in activating the Fig. 103. Effect of local P-irradiation of the preoptic nuclei of the hypothalamus on electrical activity in the sensorimotor cortex. The figures on the left of the curves indicate the time in hours and minutes from the moment the microemitter was inserted: (I) irradiation of the medial preoptic region; (11) irradiation of the preoptic nucleus. A and A ' = ISPO: oscillogram recorded f r o m the cont r a l a t e r a l cortex (A) and from the ipsilateral cortex (A'). B and B' = ECoG; oscillogram recorded from the contralateral cortex (B) and f r o m the ipsilater a l cortex (B').

212

E F F E C T O F IONIZING RADIATION

preoptic nuclei is similar to their effect on several other nerve structures (Lebedinsky et al., 1959). Analysis of the effect of X-irradiation on the parameters of the ISPO in different brain structures reveals that there a r e at least two groups of neurons in the brain. The first to react to irradiation are the neurons characterized by infraslow processes. The other neurons, which show chiefly spike activity , a r e more resistant.The sensitive neurons include, in particular, those in layers V and VI of the cortex, i.e., those regarded as the substrate of influence exerted by the subcortical structures on the metabolic "tone" of the cortex. The physicochemical properties of cortical tissue, judging from measurements of impedance in the upper layers of the cortex, s t a r t to change during the process of irradiation and go through several phases depending on the irradiation site, dose, and r a t e (Aladjalova, 1953-1956). Analysis also reveals the existence of a phase associated with the initial intensification of afferent impulses to the cortex. It appears during irradiation, regardless of the site, and gives way after a period of time, depending on the dose, to a phase of impaired trophism of the hemisphere. Direct irradiation of the head causes physicochemical changes to a r i s e in the brain some time later ("latent" period); they may be accelerated by exposing the organism to additional agents. Thereafter, processes may take place in cortical tissue that result in an increased number of f r e e ions (drop in cortical resistance). These processes are in evidence until death. However, with irradiation of the abdominal cavity, cortical resistance grows and the number of f r e e ions in tissue is sharply reduced. Thus, the physicochemical processes in the dendrite layer of the cortex undergo complex changes during and after irradiation, whether direct or remote. The data dealing with the effect of radiation on impedance in the dendrite layer of the cortex and on the ISPO in different brain structures r a i s e the question of identification of the most radiosensitive nerve elements of the cortex and suggest that investigators should direct their attention to the significance of the dendrite element of the neuron in its reaction to irradiation.

213 SUMMARY

According to Orbeli's theory, two trends can be distinguished in the evolution of living structures in their reaction to stimulation. Some structures are developing increasing chemical sensitivity and function in a close relationship with the chemical factors in the environment. Others a r e losing their chemical sensitivity and becoming more subject to control by nerve impulses. Examination of comparative physiological material shows that the first structures a r e characterized by a slow auto-oscillatory process, the second chiefly by an impulse form of activity. This general pattern appears in more complex form among the elements of the higher nerve centers. Chemically sensitive neurons that r e a c t to the appearance of humors, to local hormonal changes, can be identified in the cerebral cortex along with neurons that act mostly as a result of afferent impulses reaching them. Humoral and impulse information is synthesized in these neurons. They form the substrate of tonic regulation by the subjacent brain structures, It is these cortical neurons which, combined with the neuroglia, are characterized by the phenomenon of infraslow rhythmic potential oscillations. A s for the topography, certain of the neurons characterized by ISPO a r e located deep in the cortex while their apical dendrites reach the superficial layer. There are an unusually large number of neurons sensitive to chemical factors in the hypothalamic a r e a as well, but not i n the thalamus. Neurons of the reticular formation of the brain stem with ISPO a r e capable both of rapidly reacting to chemical changes in the medium and of rapidly returning to their original state. Integrating humoral and impulse information, the nerve elements characterized by ISPO influence the excitability of another category of neurons which exhibit mainly impulse activity. The electrical field created by the source of the ISPO may play a part in the mechanism of this influence. The f i r s t to react to ionizing radiation a r e the nerve elements with ISPO that a r e sensitive to change in the chemical factors of the environment. These include the dendrites of the neurons. The slow processes occupy an unusual place among the electrical phenomena of the brain. They reflect neurohormonal relations in which the dendrites play a major role. The slow electrical phenomena reflect the results of higher cortical functions being combined with metabolic activity aided by the mechanisms of corticohypothalamic integration. The infraslow electrical activity of the brain reflects the functioning of the slow control system of the brain, which influences the high-speed systems of the brain. Certain characteristics of the slow control s y s tem are manifested both in the biphasic intensification of infraslow activity in response to a variety of stimuli and in the phenomenon of " C fluctuations" of electrical activity in several brain structures. The existence of "C-fluctuations" in the ECoG of certain brain structures along with infraslow activity is helpful in analyzing the work of the slow control system by comparing it with the model of a regulatory system whose initial state (equilibrium) has been disrupted and then

214

SUMMARY

left alone. Regulation by the slow control system is aimed both at adjusting the system to maintain homeostasis and at including the mechanisms that actively change the p a r a m e t e r s of the system and fix these changes in the aftereffect. Thus, the slow control system may be regarded as one of the ultrastabile s y s t e m s of the body, i.e., those with a variable structure which p o s s e s s the capacity to choose for themselves the necessary working p a r a m e t e r s , rejecting unstable states and preserving those which c r e a t e new stable conditions. Such systems possess the maximum degree of adaptability. Although only the first s t e p s have been taken in the study of slow rhythmic potential oscillations in the brain, enough h a s been learned to suggest that further r e s e a r c h will throw new light on brain activity. Analysis of the phenomenon may help t o clarify certain aspects of electrical activity, provide the basis for classifying neurons by functional properties, elucidate the r o l e of the dendrites in neuron metabolism, and furnish additional approaches to gaining insight into the neurohormonal mechanisms. All this indicates that the available data will have an impact on the development of modern neurophysiological theories.

215

REFERENCES

Abel, J.J., (1924); Physiological, chemical and clinical studies on pituitary principles. Bull. Johns Hopkins Hosp., 35, 305. Abrahams, V.G., Koelle, G.B., and Smart, P., (1957); Histochemical demonstration of cholinesterases in the hypothalamus of the dog. J. Physiol., 139, 137. Achucarro, N., (1918); On the evolution of the neuroglia and specially their relation to the varicular apparatus. J. Nerv. ment. Dis., 48, 333. Adey, W.R., Merrillees, N.C.R., and Sunderland, S., (1956); The entorhinal area; behavioral, evoked potential, and histological studies of its interrelationships with brainstem regions. Brain, 79, 414. Adrian, E.D., (1936); The spread of activity in the cerebral cortex. J. Physiol., 88, 127. Adrian, E.D., (1937); Synchronized reactions in the optic ganglion of Dytiscus. J. Physiol., 91, 66. Adrian, E.D., (1941); Afferent discharge to the cerebral cortex from peripheral sense organs. J. Physiol., 100, 159. Adrian, R.H., (1956); The effect of internal and external potassium concentration on the membrane potential of frog muscle. J. Physiol., 133, 631. Adrianov, O.S., (1958); The connections and functions of the unspecific reticular formation of the optic lobe. Proc. 8th Conf, on higher nervous Activity, 1, 5. Agafonov, V.G., (1956); The inhibitory effect of aminazine on the central projection of pain stimuli. Korsakov J. Neuropathol. Psychiat., 56, 64. Aladjalova, N.A., (1950a); Electrical parameters of muscles with different functions. Lectures Acad. Sci. USSR, 71, 45. Aladjalova, N.A., (1950b); Sympathic influence on the electrical parameters of striated muscle and the rhythmic nature of their fluctuations. Lectures Acad. Sci. USSR, 73, 73. Aladjalova, N.A., (1953a); Research on physico-chemical phenomena in the living brain by recording of the electrical parameters. Proc. 16th Conf. on higher nervous Activity, 7. Aladjalova, N.A., (1953b); Early physico-chemical changes in the cerebral cortex evoked by irradiation of head and trunk. Proc. Conf. on the Biochemistry of the nervous System. Kiev, 3. Aladjalova, N.A., (1953~);On the application of the method of "dielectric loss" in relation to abnormal absorption in food quality testing. Biophysics in Agriculture. Moscow. Aladjalova, N.A., (1954a); Electrical parameters of the cerebral cortex. Lectures Acad. Sci. USSR, 94, 1053. Aladjalova, N.A., (1954b); Electrical parameters of the cerebral cortex and their dependance on the physico-chemical aspects of cerebral activity. Proc. Conf. on Nature and Research of bioelectrical Potentials, p. 5. Aladjalova, N.A., (1955a); Research on the physico-chemical properties of the living cerebral cortex under various functional circumstances. Rep. Inst. Biophysics, USSR Acad. Sci., 1, 14. Aladjalova, N.A., (1955b); The method of synchronized recording of the resistance and capacitance components of electrical impedance in rapidly changing processes. Rep. Inst. of Biophysics, USSR Acad. Sci., 1, 298. Aladjalova, N.A., (1956a); Dynamics of early ionic movements in the cerebral cortex during localized x-irradiation of gastric a r e a and head. J. Biophysics, 1, 1949. Aladjalova, N.A., (1956b); Infraslow rhythmical changes in the electrical potentials of the brain. J. Biophysics, 1, 127.

216

REFERENCES

Aladjalova, N.A., (1956~);Changes in infraslow rhythmical potential fluctuations in the brain under influence of ionizing radiation. J. Biophysics, 1, 642. Aladjalova, N.A., (1957a); Infraslow rhythms in the potential fluctuations of the cerebral cortex and the influencing factors. Proc. Conf. on Electrophysiology of the central nervous System, Leningrad. p. 8. Aladjalova, N.A., (1957b); Infraslow rhythms in the electrical potential fluctuations of the cerebral cortex and hypothalamus in radiation injury. Proc. Conf. on the Technology of radioactive and stabile Isotopes and Radiation in Agriculture and Science. Aladjalova, N.A., (1957~);Infraslow oscillations of the steady potential of the cerebral cortex. Nature, 179, 957. Aladjalova, N.A., (1958a); Pharmacological influences on the infraslow potential fluctuations of the cerebral cortex as shown in the ECoG. USSR J. Physiol., 44, 793. Aladjalova, N.A., (1958b); The influence of hypothalamus and reticular formation on infraslow potential fluctuations of the cerebral cortex. Proc. Conf. on Electrophysiology of the central nervous System, 9. Aladjalova, N.A., (1958~);Infraslow rhythms in the electrical potential fluctuations of the cerebral cortex and hypothalamus during irradiation. Proc. Conf. on Technology of radioactive and stabile Isotopes in Agriculture and Science, 184. Aladjalova, N.A., (1959a); Research on the electrical properties of various tissues in the living animal. Proc. Conf. on Radioelectronics in Medicine and Biology, 2. Aladjalova, N.A., (1959b); Infraslow potential fluctuations in the cerebral cortex and some subcortical structures. Proc. 9th Congr. of Physiologists, 21. Aladjalova, N.A., (1960a); Some aspects of biophysics of the cerebral cortcx. Physicochemical and structural Principles in Life and its Phenomena. Moscow. Aladjalova, N.A., and Koltsova, A.V., (1958); On infraslow rhythmic potential fluctuations in hypothalamic and thalamic nuclei. Bull. exp. Biol. Med., 10, 3. Aladjalova, N.A., and Koltsova, A.V., (1962); Rhythms of electrical activity in various structures of the brain. Lectures USSR Acad. Sci., 142, 1. Aladjalova, N.A., and Koshtoyants, O.K., (1957); Microelectrode research on the pseudo-stabile potential and its infraslow oscillations on different levels of the cerebral cortex. Biophysics, 2, 327. Aladjalova, N.A., and Koshtoyants, O.K., (1960); Infraslow rhythmic potential fluctuations in isolated slabs of the cerebral cortex. USSR J. Physiol., 46, 12. Aladjalova, N.A., and Maslov, N.M., (1957); Range of abnormal dielectrical losses of striated and smooth muscle. Lectures USSR Acad. Sci., 115, 407. Aladjalova, N.A., and Mertsalova, S.N., (1954); Differences in changes and electrical properties of smooth muscle in rapid and tonic contraction. Lectures USSR Acad. Sci., 94, 213. Aladjalova, N.A., and Mertsalova, S.N., (1958); Range of abnormal dielectrical losses of smooth muscle in occlusive tonus. Biophysics, 3, 23. Albe-Fessard, D., (1951); Modifications de l'activitb des organes 6lectriques par des courants d'origine extbrieure. Arch. Sci. Physiol., 5, 45. Aleksandrovskaya, M.M., (1958); Data on the influence of ionizing radiation on the structure of the central nervous system in animals. Lectures Inst. higher nervous Activity, 4, 21. Aleksanyan, A.M., and Demirchoglyan, G.G., (1955); On slow oscillations of electrical potentials in the visual analyser. Lectures Armenian Acad. Sci., 8, 67. Aleshin, B.V., (1960); On the significance of the hypothalamus for the control of the anterior lobe of the hypophysis. Progr. contemp. Biol., 49, 115. Amin, A.N., Crawford, T.B.B., and Gaddum, J.H.,(1954); The distribution of substance P and S-hydroxytryptamine in the central nervous system of the frog. J. Physiol., 126, 596. Anokhin, P.K., (1949); Important problems in the research on higher nervous activity. Problems of higher nervous Activity, 105. ' lar formation in the eliciting of Anokhin, P.K., (1956); On the r ol e o-f

REFERENCES

217

unconditioned stimuli in the cerebral cortex. Proc. 20th Int. Congr. of Physiologists, Brussels, p. 151. Anokhin, P.K., (1958a); Internal Inhibition as a physiological Problem. Moscow. Anokhin, P.K., (1958b); Recent data on the reciprocity of cortex and subcortical structures. P r o c . 18th Conf. on higher nervous Activity, 1, 10. Araki, T., and Otani, T., (1955); Response of single motoneurones to direct s t i m ulation in toad's spinal cord. J. Neurophysiol., 18, 472. Arduini, A., Mancia, M., and Mechelse, K., (1957); Slow potential changes elicited in the cerebral cortex by sensory and reticular stimulation. Arch. ital. Biol., 95, 127. Arkhangelsky, V.N., (1958); The intermediate substance of the retina and its significance in metabolic processes and visual functions in man. Lectures USSR Acad. Sci., 9, 21. Artemyev, V.V., (1951); Electrical reaction of the cerebral cortex to acoustic stimulation in anesthesized and unanesthesized animals. USSR J. Physiol., 37, 688. Artemyev, V.V., and Babskii, E.B., (1949); Electrophysiological analysis of the effect of acetylcholine on brain nuclei. USSR J. Physiol., 35, 623. Arvanitaki, A., (1938); Les Variations graduees de la Polarisation des Systsmes excitables. Paris. Arvanitaki, A,, (1939a); Caractsres de l'activite 6lectrique gradu6e e t locale de l'axone is016 d e Skpia. C.R. SOC.Biol. Paris, 130, 424. Arvanitaki, A., (1939b); Local responses in decalcified giant axon. Arch. Internat. Physiol., 49, 206. Arvanitaki, A., (1939~);Recherches s u r la response oscillatoire locale de l'axone g6ant is016 de SBpia. Arch. Internat. Physiol., 49, 209. Arvanitaki, A., (1942); Ephapse transmission; local potentials, S6pia giant axons. J. Neurophysiol., 5, 89. Arvanitaki, A., and Chalazonitis, N., (1956); Biopotentiels neuroniques Q 1'Bchelle intracellulaire. Bull. Inst. OcBanographique, 1079. Aseyev, B.P., (1954); Phase Relations in Radiotechnics. Moscow. Ashby, W. Ross, (1954); Design. for a Brain. London. Ashby, W. Ross,(1960); T h e b r a i n a s regulator. Nature, 186, 4722, 413. Asratyan, E.A., (1930); Influence of extirpation of the superior cervical ganglia on the conditioned feeding reflexes in the dog. Rec. biol. Sci., 3, 243. Asratyan, E.A., (1959); A Lecture on some Problems of Neurophysiology. Moscow. Asratyan, E.A., and Aleksanyan, A.M., (1933); Conditioned reflex activity in brains with ablation of the cerebral cortex and diencephalon. USSR J. Physiol., 16, 6. Auer, J., (1956); Terminal degeneration in the diencephalon after ablation of frontal cortex in the cat. J. Anat., 90, 30. Barnes, G . , (1955); The behavior of Anodonta cygnea L. and its neuro-physiological basis. J. exp. Biol., 32, 158. Barron, G., (1955); The effect of ionizing radiation on biologically important systems. Radiobiology . Moscow. Barry, J., (1954a); Neuros6cr6tion hypothalamique de substance colloide chez le rat blanc et la s o u r i s blanche. C. R. Soc. Biol. Paris, 148, 561. Barry, J., (1954b); Neurocrinie de substance neuros6cr6toire d'origine hypothalamique. C. R. SOC.Biol., 148, 1459. Barry, J., (1954~);Sur l'existence de cellules acidophiles de type neuros6cr6toire au niveau de l'hypothalamus chez la cobaye. C. R. SOC.Biol. Paris, 148, 133. Barry, J., (1956); De l'existence probable de synapses interneuronales de type "neuros6cr6toire". Progr. in Neurobiol., 36. Beck, A., (1890); Oznacznie lokalizacyi w m6zgu i rdzeniu z a pomoca zjawisk elektrycznych. Thesis for the University of Jagiellanski. Krakov. Bekhterev, V.M., (1896); The development of neurons in the cerebral cortex of man. Sel. Works, Moscow, 76. Beritashvili, I.S., (1949); On the origin of infraslow potentials of the brain. Proc. Gagra Symp., Tbilisi, 1, 209.

218

REFERENCES

Beritashvili, I.S., (1953); Morphological and physiological data on neural connections in the cerebral cortex. Proc. 13th Session of the Section Biology and Medicine of the Gruzian Academy of Sciences, Tbilisi, 3. Beritov, I.S., (1937); On the conductivity of the central nervous system of homostatic animals. Rep. Beritashvili Inst. Physiol., 3, 37. Beritov, I.S., (1941); On the processes of excitation, inhibition and facilitation in the spinal cord. Rep. Beritashvili Inst. Physiol., 4, 1. Beritov, I.S., (1948a); The nature of spontaneous electrical activity of the central nervous system. Rep. Beritashvili Inst. Physiol., 5, 215. Beritov, L.S., (1948b); General neuro- and muscle Physiology.' Moscow-Leningrad. Beritov, I.S., (1957); On the essential F o r m s of nervous and psycho-nervous Activity. Moscow - Leningr ad. Bethe, A., (1897); Das Nervensystem von Carcinus maenas. Arch. mikr. Anat., 50, 51. Betz, E.H., (1956); Contribution 2 1'Etude du Syndrome endocrinien. Paris. Biryukov, D.A., (1957); On the mechanism of penetrating radiation'in the nervous system. USSR J. Physiol., 47, 637. Bishop, G.H., (1949); Potential phenomena in thalamus and cortex. Electroenceph. clin. Neurophysiol., 1, 421. Bishop, G.H., and Clare, M.H., (1952); Sites of origin of electrical potentials in striate cortex. J. Neurophysiol., 15, 201. Bishop, G.H., and O'Leary, J.L., (1950); Effects of polarizing currents on cell potentials and their significance in the interpretation of central nervous activity. Electroenceph. clin. Neurophysiol., 2, 401. Blair, E.A., and Erlanger, J., (1936); Temporal summation in peripheral nerve fibres. Amer. J. Physiol., 117, 355. Bok, S.T., (1959);-The synapses in the cerebral cortex. Progr. in Neurobiol., 13. Bonnet, V., and Bremer, F., (1937); Action du potassium, du calcium e t de l'acetylcholine sur les activitks Blectriques, spontankes et provoqukes, de l'kcorce ckrkbrale, C. R. SOC.Biol., 126, 1271. Bonvallet, M., Dell, P., and Hiebel, G., (1954); Tonus sympathique et activitk 6lectrique corticale. Electroenceph. clin. Neurophysiol., 6, 119. Bonvallet, M., Dell, P., and Hugelin, A., (1954); Influence de l'adrknaline s u r l e contrOle rkticulaire des activitks corticales e t spinales. J. Physiol. (Paris), 46, 262. Bonvallet, M., Hugelin, A., and Dell, P., (1956); Milieu interieur et activitk automatique des cellules reticulaires mksenckphaliques. J. Physiol., (Paris), 48, 403. Borison, H.L., and Wang, S.C., (1953); The physiology and pharmacology of vomiting. Pharmacol. Rev., 5, 193. Bozler, E., (1935); The change of alternating current impedance of muscle produced by contraction. J. cell. comp. Physiol., 6, 217. Bradley, K.,Easton, D.M., and Eccles, J.C., (1953); An investigation of primary or direct inhibition. J. Physiol., 122, 474. Bradley, P.B., and Elkes, J., (1957); The effects of some drugs on the electrical activity of the brain. Brain, 80, 1, 177. Bremer, F., (1958); Analysis of processes of stimulation in the cerebral cortex. Proc. int. Symp. on the EEG of higher nervous Activity. Moscow. Brock, L.G.,Coombs, J.S., and Eccles, J.C., (1952); The recording of potentials from motoneurones with an intracellular electrode. J. Physiol., 117, 431. Brodal, A., (1957); The reticular formation of the brain stem. Anatomical Aspects and functional Correlation. Brodie, B.B., and Shore, P.A., (1957). A concept for a role of serotonin and norepinephrine as chemical mediators in the brain. Ann. N.Y. Acad. Sci., 66, 631. 'Broqckhart, J.M., Arduini, A., Mancia, M., and Moruzzi, G., (1958); Thalamocsrtical relations as revealed by induced slow potential changes. J. Neurophysiol., 21,499.

REFERENCES

219

Brown, G.L., and Feldberg, W., (1936); The acetylcholine metabolism of a sympathetic ganglion. J. Physiol., 88, 265. Bulbring, E., (1957); Changes in configuration of spontaneously discharged spike potentials f r om smooth muscle of the guinea-pig's Taenia coli. J. Physiol., 135, 412, Bulbring, E . , and Burn, J.H., (1941); Observations bearing on synaptic transmission by acetylcholine in the spinal cord. J. Physiol., 100, 337. Biilbring, E., Burn, J., and Shelley, H.,(1953); Acetylcholine and ciliary movement in the gill plates of Mytilus edulis. Proc. roy. Soc. (B). 141, 445. Biilbring, E., and Lin, R.C.H., (1957); J. Physiol., 13&, 12. Bureg, J., (1954a); Direct potential difference between the cerebral hemispheres during the depression of EEG activity in anesthesized and nonanesthesized r a t s . Physiol. Bohemoslovenica, 3, 288. Bureg, J., (1954b); On the question of electrotonic mechanisms in the activity of the central nervous system. The production of spreading depression of EEG activity by electrotonus. Physiol. Bohemoslovenica, 3, 272. Burex, J., (1955); The relation between the depression of EEG activity brought about by electrotonus and the narcotic effect of a direct current. Physiol. Bohemoslovenica, 4, 98. Bure3, J., (1957); The ontogenetic development of steady potential differences i n the cerebral cortex in animals. Electroenceph. clin. Neurophysiol., 9, 121. Bureg, J., and Burexova, O., (1956); The question of ionic antagonism in spreading depression. Physiol. Bohemoslovenica, 5, 2 , 195. Bureg, J., and Buresova, O., (1958); Application of reversible depolarization of the cer eb r a l cortex (by means of a spreading depression) in the research of conditioned reflexes. Proc. int. Conf. on the EEG of higher nervous Activity. Moscow. Burez, J., and Koshtoyants, K.S., (1955); The role of sulphydryl groups in the propagation of a spreading depression in the cerebral cortex. Lectures USSR Acad. Sci., 105, 5. Burns, B.D., (1951a); Some properties of the c a t ' s isolated cerebral cortex. J. Physiol., 111, 50. Burns, B.D., (1951b); Some properties of isolated cerebral cortex in the unanesthesized cat. J. Physiol., 112, 156. Burns, B.D., (1954); The production of afterburst in isolated unanesthesized cerebral cortex. J. Physiol., 125, 427. Burns, B.D., (1958); The mammalian cerebral Cortex. London. Burns, B.D., and Grafstein, B., (1952); The function and structure of some neurones in the cat's cer ebr al cortex. J. Physiol., 118, 412. Burns, B.D., Grafstein, B., and Olszewski, J., (1957); Identification of neurones giving bur s t response in isolated cerebral cortex. J. Neurophysiol., 20, 200. Burr, H.S., (1944); The meaning of bio-electrical potentials. Yale J. Biol. Med., 16, 353. Burr, H.S.,(1950); Potential gradients in living systems and their measurements. Med. Physiol., 2, 90. Burr, H.S.,and Mauro, A., (1949); Electrostatic fields of the sciatic nerve in the frog. Yale J. Biol. Med., 21, 455. BYZOV, A.L., (1958); Resistance and capacitance in different layers of the frog retina. Biophysics, 3, 658. BYZOV, A.L., and Bongardt, M.M., (1959); A cathode follower for research with microelectrodes. USSR J. Physiol., 45, 111. Cajal, S. R m 6 n y, (1925); Contribucion a1 conocimiento de la neuroglia del cerebro humano. Trab. Lab. Invest. biol. Univ. Madr., 11, 255. Cajal, S. RamBn y, (1928); Degeneration and regeneration of the nervous system. Trans. R.M. May, London. Cajal, S. Ramdn y, (1934); Les preuves objectives de l'unite anatomique des cellules nerveuses. Trab. Lab. Invest. biol. Univ. Madr., 29, 1.

220

REFERENCES

Caspers, H., (1959); Uber die Beziehungen zwischen Dendritenpotential und Gleichs w u n g an d e r Hirnrinde. Pflligers Archiv, 269, 1957. Castillo, J. del, and Katz, B., (1956); Biophysical aspects of neuromuscular transmission. Pr ogr . Biophys., 6, 121. Caton, R., (1891); Die Strbme d e s Centralnervensystems. Obl. Physiol., 4, 785. Cavanagh, J.B., Thompson, R.H.S., and Webster, G.R., (1954); Localization of pseudocholinesterase activity in nervous tissue. Quart. J. exp. Physiol., 39, 185. Chagovyets, V.Y., (1898); The application of the dissociation theory of Arrhenius of electrolyte solutions in electrophysiology. Physiology of the nervous System. Moscow, 1952, p. 594. Chailahyan, L., (1958); Interrelations of Action Potentials and Changes of Conductivity of myelinated Nerve. Thesis, Moscow State University. Chang, H.T., (1951a); Dendritic potential of cortical neurons produced by direct electrical stimulation of the cer ebr al cortex. J. Neurophysiol., 14, 1. Chang, H.T., (1951b); Changes in excitability of cerebral cortex following single electric shock applied to cortical surface. J. Neurophysiol., 14, 95. Chang, H.T., (1952a); Cortical response to stimulation of lateral geniculate body and the potentiation thereof by continuous illumination of retina. J. Neurophysiol., 15, 1, 5. Chang, H.T., (1952b); Cortical and spinal neurons. Cortical neurons with particular reference to the apical dendrites. Cold Spr. Harb. Symp. quant. Biol., 17, 189. Chang, H.T., (1953a); Cortical response to activity of callosal neurons. J. Neurophysiol., 16, 117. Chang, H.T., (1953b); Interaction of evoked cortical potentials. J. Neurophysiol., 16, 133. Chang, H.T., (1958); Some observations on the change in excitability of cortical and subcortical neurons and their role in eliciting a conditioned reflex. Proc. int. Colloquium on the EEG of higher nervous Activity. Moscow. Chang, J.J., and Hild, W., (1959); Contractile responses to electrical stimulation of glial cells from the mammalian central nervous system cultivated in vitro. J. cell. comp. Physiol., 53, 139. Clare, M.H., and Bishop, G.H., (1954); Responses from association area secondarily activated from optic cortex. J. Neurophysiol., 17, 271. Clare, M.H., and Bishop, G.H., (1955a); Properties of dendrites; apical dendrites of the cat cortex. Electroenceph. clin. Neurophysiol., 7, 85. Clare, M.H., and Bishop, G.H., (1955b); Dendritic circuits: the properties of cortical paths involving dendrites. Amer. J. Psychiat., III, 818. Clare, M.H., and Bishop, G.H., (1956); Potential wave mechanisms in cat cortex. Electroenceph. cl'in. Neurophysiol., 8, 583. Coldwell, P.S., and Keynes, R.D., (1957); The utilization of phosphate bound energy for sodium extrusion from giant axons. J. Physiol., 137, 12. Cole, K., (1928); Electric impedance of suspensions of spheres. J. gen. Physiol., 12, 29. Cole, K., (1932); Electric phase angle of cell membranes. J. gen. Physiol., 15, 641. Cole, K., (1933); Electric conductance of biological systems. Cold Spr. Harb. Symp. quant. Biol., 1, 107. Cole, K., (1934); Alternating current conductance and direct current excitation of nerve. Science, 79, 164. Cole, K., (1941); Rectification and inductance of the squid giant axon. J. gen Physiol., 25, 981. Cole, K., (1949); Some physical aspects of bioelectric phenomena. Proc. nat. Acad. Sci., 35, 558. Cole, K., (1955); Ions potential and the nerve impulse. Electrochemistry in Biology and Medicine. London, p. 121. Cole, K., and Backer, O., (1941a); Transverse electric impedance of the squid giant axon. J. gen. Physiol., 25, 535. Cole, K., and Backer, O., (1941b); Longitudinal electric impedance of the squid giant

REFERENCES

22 1

axon. J. gen. Physiol., 25, 771. Cole, K., and Curtis, H.,(1937); Wheatstone bridge and electrolytic resistor for impedance measurements over a wide frequency range. Rev. Sci. Instrum., 8, 333. Cole, K., and Curtis, H., (1938); Electxic impedance of Nitella during activity. J. gen. Physiol., 22, 37. Cole, K., and Curtis, H., (1939); Electric impedance of the squid giant axon during activity. J. gen. Physiol., 22, 649. Cole, K., and Curtis, H.,(1941); Membrane potential of the squid axon during current flow. J. gen. Physiol., 24, 551. Cole, K., and Curtis, H., (1950); Bioelectricity. Medical physics, 2, 82. Colfer, H.F., and Essex, H.E., (1947); The distribution of total electrolyte. K and Na in the cerebral cortex in relation to experimental convulsions. Am. J. Physiol., 150, 27. Collin, R., (1956); Les ph6nomenes hydrenc6phalocrines comme sources accessoires hormonales du liquide cbphalio-rachidien. P r o g r e s s in Neurobiology. Proc. first int. meeting neurobiologists. Amsterdam, p. 172. Collin, R., and Barry, J., (1954); Hydrenc6phalocrinie neuros6cr6toire dans l e ventricule dienc6phalique chez le crapaud. Ann. Endocrin., 15, 533. Coombs, J.S., Curtis, D.R., and Eccles, J.C., (1956); Time courses of motoneuronal responses. Nature, 178, 1049. Coombs, J.S., Eccles, J.C., and Fatt, P., (1955a); The electrical properties of the motoneurone membrane. J. Physiol. , 130, 291. Coombs, J.S., Eccles, J.C., and Fatt, P., (195513); The specific ionic conductances and the ionic movements a c r o s s the motoneuronal membrane that produce the inhibitory postsynaptic potential. J. Physiol., 130, 326. Coorvoisier, S.J., Fournel, J., Ducrot, R., Kolsky, M., and Koetshut, P., (1953); Propri6t6s pharmacodynamiques du chlorhydrate de chloro-3 (dimethyl-amino-3propy1)-10-phenothiazine.Arch. int. Pharmacodyn. , 92, 305. Dale, H.H., (1935); Pharmacology and nerve endings. Proc. roy. SOC.Med., 28, 319. Dale, H.H., (1954); The beginning and the prospects of neurohumoral transmission. Pharmacol. Rev., 6, 7. Danilevskii, V.Y., (1875); Research on the Physiology of the Brain. Thesis, Moscow. Danilevskii, V.Y., (1891); Electrical phenomena in the brain. Earliest Experiments on the EEG in Russia. Moscow (1949), 77. Debay, P., (1931); Polar Molecules, Moscow-Leningrad. Demin, N.N., (1952); The biochemical Activity of Acetylcholine. Thesis, Moscow. Demirchoglyan, G.G., Adunts, G.T., and Avakyan, T.M., (1957); Effect of radiation on the functional state of the retina of the eye. Proc. nat. Conf. on medical Radiology , 88. Dempsey, F.W., and Luse, S., (1958); Fine structure of the neuropil in relation to neuroglia cells. Symposium Biology of neuroglia. Springfield, p. 99. DeRobertis, E., and Bennet, H.S., (1955); Some features of fine structure of cytoplasm of cells in the earthworm nerve cord. Symp. on fine Structure of Cells, p. 261. Deryabin, V.S., (1946); Influence of lesion of the optical thalamus and the hypothalamic a r e a on higher nervous activity. USSR J. Physiol., 32, 533. Desmedt, J.E., and Franken, L., (1957); Facteurs neurohumoraux de l'activit6 6lectrique de 1'6corce c6r6brale. F r e e communications. B. Neurophysiology. Int. Neurol. Congr., B r w e l l e s , p. 148. Desmedt, J.E., and L a Grutta, G., (1955); Control of brain potentials by pseudocholinesterase. J. Physiol., 129, 46. Desmedt, J.E., and La Grutta, G., (1957); The effect of selective inhibition of pseudocholinesterase on the spontaneous activity and evoked potential of the cortex. J. Physiol., 136, 1, 20. Dixon, K.C., (1949a); Anaerobic leakage of potassium from brain. Biochem. J., 44, 187.

222

REFERENCES

Dixon, K.C., (1949b); Action of potassium ions of brain metabolism. J. Phys
REFERENCES

223

F e s s ar d , A., (1936); Recherches s u r l'activit6 rythmique des nerfs isoles. Paris. F e s s ar d , A., (1954); Mechanisms of nervous integration and conscious experience Symp. Brain mechanisms and consciousness, 200. F e s s ar d , A., (1956); For m es e t car act eres g e n e r a w de l'excitation neuronique. Proc. XX int. Congr. Physiol. B r w e l l e s , p. 35. Florey, E., and McLennan, H., (1955); The rel ease of an inhibitory substance from mammalian brain, and its effect on peripheral synaptic transmission. J. Physiol., 129, 384. Ford, D.H., and Kantonnis, N.I. , (1957); The localization of neurosecretory structures and pathways in the male albino rabbit. J. Comp. Neurology, 108, 91. Frank, G.M., (1956); On the structural and physicochemical processes in muscle contraction. Problems of contemporary Physiology of the nervous and muscle Systems. Tbilisi, p. 475. Frank, G.M., (1960); Mobility of cellular structures and the problems of regulation of biological processes. Physicochemical and structural Principles of Life and its Phenomena. Frank, G.M., Aladjalova, N.A., and Snezhko, A.D., (1959); Some problems of the biophysical analysis of radiobiological effects. Proc. 2nd Conf. on the peaceful Use of atomic Energy. Section Radiobiology and Radiation Medicine, 123. Frank, G.M., Aladjalova, N.A., Veize, L.G., and Snezhko, A.D., (1955); The early reactions of the organism to X-irradiation dependent on the site of influence. Proc. 8th Congr. of Physiologists, Biochemists and Pharmacologists, 643. Frank, K., and Fuortes, M.G.F., (1955); Potentials recorded from the spinal cord with microelectrodes. J. Physiol., 130, 625. Frey-Wyssling, A. , (1948); Submicroscopic Morphology of Protoplasm. Freygang, Jr., W.H., and Landau, W.M., (1955); Some relations between resistivity and electrical activity in the cer ebr al cortex of the cat. J. cell. comp. Physiol., 45, 3, 377. Fricke, H., (1931); Electrical conductivity and capacity of disperse system. Physics, 1, 106. Fricke, H., (1933); The electric impedance of suspensions of biological cells. Cold Spr. Harb. Symp. quant. Biol., 1, 117. Fricke, H., and Morse, S., (1925); The electric resistance and capacity of blood for frequencies between 800 and 4 i million cycles. J. gen. Physiol., 9, 153. Frumkin, A.N., Bagotskii, V.S., Ioffe, Z.A., and Kabanov, B.N., (1952); Kinetics of Electrode Processes. Moscow. Galambos, R., (1961); A glia-neural theory of brain function. Proc. Nat. Acad. Sci. USA, 47, 129. Galkin, V.S., (1953); A short Study of the Theory of Narcosis. Leningrad. Gellhorn, E., (1954); Physiological processes related t o consciousness and perception. Brain, 77, 401. Gellhorn, E., Nakao, H., and Redgate, E.S., (1956); The influence of lesions in the anterior and pobtefioi hypothalamus on tonic and phasic autonomic reactions. J. Physiol., 131, 402. Genes, S.G., (1955); The nervous System and internal Secretion. Moscow. Gerard, R.W., (1941); Interaction of neurones. Ohio J. Sci., 41, 160. Gerard, R.W., and Libet, B., (1940); The control of normal and convulsive brain potentials. Amer. J. Psychiat., 96, 1125. Gershun, G.V., and Tonkikh, A.V., (1949); Electrical manifestations of the activity of various par t s of the nervous system in the waking and sleeping cat. Rep. Pavlov Inst. Physiol., 3, 11. Giacobini, E., (1959); The 'distribution and localization of cholinesterase in nerve cells. Acta physiol. S c a d . , 45, 156. Glees, P., (1955); Neuroglia, Morphology and Function. Oxford. Goldring, S., and O'Leary, J.L., (1951a); Experimentally derived correlates between EEG and steady cortical potential. J. Neurophysiol., 14, 275.

224

REFERENCES

Goldring, S., and O'Leary, J.L., (1951b); Summation of certain enduring sequelae of cortical activation in the rabbit. Electroenceph. clin. Neurophysiol., 3, 329. Goldring, S., and O'Leary, J.L., (1954); Correlation between steady transcortical potential and evoked response. I. Alterations in somatic receiving area and induced by veratrine, strychnine, KC1, and novocaine. Electroenceph. clin. Neurophysiol., 6, 189. Goldring, S., and O'Leary, J . L . , (1957); Cortical d.c. changes incident to midline thalamic stimulation. Electroenceph. clin. Neurophysiol., 9, 577. Goldring, S., O'Leary, J . , and Huang, S.H., (1958); Experimental modification of dendritic and recruiting processes and their d.c. after-effects. Electroenceph. clin. Neurophysiol., 10, 663. Goldring, S., Ulett, G., O'Leary, J., and Greditzer, A., (1950); Initial survey of slow potential changes obtained under resting conditions and incident to convulsive therapy. Electroenceph. clin. Neurophysiol., 2, 297. Goldstein, B.I., (1959); Biochemical principles of the mechanism of action of hormons. Action Mechanisms of Hormons. Kiev, p. 15. Gorizontov, P.D., (1959); Pathophysiology of radiation injuries. J. radiation Med., 80. Gorizontov, P.D., (1959a); The problem of the pathogenesis of acute radiation sickness and its pathophysiological aspect. Proc. 3rd int. Conf. on the peaceful Application of atomic Energy. Section Radiobiology and Medicine, 43. Grafstein, B., (1956a); Mechanism of spreading cortical depression. J. Neurophysiol., 19, 154. Grafstein, B., (1956b); Locus of propagation of spreading cortical depression. J. Neurophysiol., 19, 308. Graschenkov, N.I., (1959); On the nervous and humoral mechanisms of corticosubcortical interaction (the r o l e of the hypothalamus and its connections). Proc. 9th Congr. nat. SOC.of Physiologists, Biochemists and Pharmacologists, MoscowMinsk, 1, 168. Gray, J., (1924); The relation of ciliary activity to oxygen consumption. Proc. roy. SOC.B., 96, 95. Green, J.D., and Arduini, A.A., (1954); Hippocampal electrical activity in arousal. J. Neurophysiol., 17, 533. Green, J.D., and Naquet, R., (1957); Etude de la propagation locale et 5 distance de decharges epileptiques. I. Congr. int. Sci. Neurol. IV Congr. EEG. Bruxelles, p. 187. Grigoryev, Y.G., (1958); Contributions to the Study of Reactions of the human cent r a l nervous System on penetrating Radiation. Moscow. Grindel, O.M., Maiorchik, V.E., Puchinskaya, L.M., Rusinov, V.S., and Sokolova, A.A., (1959); Electrocortical responses to the existence of stimulating foci in various parts of the human brain stem. Proc. 9th Congr. Physiologists, Biochemists and Pharmacologists, Moscow-Minsk, 1, 183. Grunbaum, A.S.F., and Sherington. C., (1903): Observation on the physiology of c e r ebral cortex of anthropoid apes. Proc. roy. SOC.,London, 72, 152. Grundfest, H., (1952); Mechanism and properties of bioelectric potentials. Modern Trends in Physiology and Biochemistry, p. 193. Grundfest, H., (1956); Some properties of excitable tissue. Fifth Conf. Nerve Impulse. New York. Grundfest, H., (1957a); Electrical inexcitability of synapses and some consequences in the central nervous system. Physiol. Rev.? 37, 3, 337. Grundfest, H., (1957b); General problems of drug action on bioelectric phenomena. Ann. N.Y. Acad. Sci., 663. 537. Grundfest, H., ( 1 9 5 7 ~ )Excitation ; at synapses. J. Neurophysiol., 20, 316. Grundfest, H., (1958); An electrophysiological basis for cone vision in fish. Arch. ital. Biol., 96: 2, 135.

REFEIiENCES

225

Grundfest, H., and Purpura, D.P., (1956); The postsynaptic potentials of cat cortical dendrites. P r o c . XX int. Physiol. Congr., 374. Guillemin, R., (1957); Centrally acting drugs and pituitary adrenal responses to s t r e s s . Brain Mechanisms and Drug Action. Oxford, p. 99. Guillemin, R., Hearn, W.R., Cheek, W.R., and Housholder, D.E., (1956); Isolation from hypothalamus of substance which stimulates r e l e a s e of ACTH in vitro. Fed. Proc., 15, 84. Gulyayev, P.I., (1956); Reflection of some Properties of Inihibition during Sleep on the electrical P r o c e s s e s of the human cerebral Cortex. Theqis, Leningrad University. Hagiwara, S., and Bullock, T., (1957); Intracellular potential i n pacemaker and integrative neurons of the lobster cardiac ganglion. J. cell. comp. Physiol., 50, 1, 25. Hagiwara, S., and Tasaki, I., (1958); A study of the mechanism of impulse t r a n s m i s sion a c r o s s the giant synapse of the squid. J. Physiol., 143, 114. Harris, G.W., (1955); Neural Control of the pituitary Gland. London. Harris, G.W., (1960); Central control of the pituitary secretion. Handbook of Physiology, II. Washington. Hebb, C.O., and Smallman, B.N., (1956); Intracellular distribution of choline acetylase. J. Physiol., 134, 385. Heilbrunn, L., (1957); Dynamics of living Protoplasm. Moscow. Henry, C.E., and Scoville, W.B., (1952); Suppression-burst activity from isolated cerebral cortex in man. Electroenceph. clin. Neurophysiol., 4, 1-2. Hermann, L., (1879); Handbuch der Physiologie. Hess, W.R., (1946); Experimentelle Darstellung des mesencephalen Blickzentrums, seiner Struktur und Funktionsweise. Helv. physiol. Acta, 4, 3. Hess, W.R., (1949); Das Zwischenhirn. Basel. Hicks, S., and Montgomery, Ph., (1952); Effects of acute radiation of the adult mammalian central nervous system. Proc. SOC.exp. Biol. Med., 80, 1, 15. Hild, W., Chang, J.J., and Tasaki, I., (1958); Electrical responses of astrocytic glia from the mammalian central nervous system cultivated in vitro. Experientia, 14, 6 . Hiiber, R., (1910); Eine Methode die elektrische Zeitfmigkeit im Inneren von Zellen zu messen. Pfl. Arch., 133, 237. Hodgkin, A.L., (1951); The ionic basis of electrical activity in nerve and muscle. Biol. Rev., 26, 339. Hodgkin, A.L., (1957); Ionic movements and electrical activity in giant nerve fibres. Proc. roy. SOC.(B) London, 148, 1. Hodgkin, A.L., and Keynes, R.D., (1953); The mobility and diffusion coefficient of potassium in giant axons from Sepia. J. Physiol., 119, 513. Holton, F.A., and Holton, P., (1954); The capillary dilator substances in dry powders of spinal roots: a possible r o l e of adenosine triphosphate in chemical transmission from nerve endings. J. Physiol., 125, 124. Horstmann, E., (1954); Uber kernspezifische Synapsenformen im Mittelhirn von Knochenfischen. Z. Zellforsch., 40, 139. Huxley, A.F., (1954); Electrical processes in nerve conduction. Symp. Iontransport a c r o s s Membranes. New York, p. 23. Hyde'n, H., (1943); Protein metabolism i n the nerve cell during growth and function. Acta physiol. Scand., 28, 44. Ilyina, A.I., and Tonkikh, A.V., (1957); New data on a neurohumoral link in vascular reaction. USSR J. Physiol., 43, 3. Ingram W.R., (1960); Central autonomic mechanisms. Handbook of Physiology, 11, p. 951. Washington. Ingvar, D.H., (1955a); Reproduction of the three per second spike and wave EEG pattern by subcortical electrical stimulation in cats. Acta physiol. Scand., 33, 2-3, 137. Ingvar , D.H., (1955b); Electrical activity of isolated cortex in the unanesthesized cat with intact brain stem. Acta physiol. Scand., 33, 2-3, 151. Ingvar, D.H., ( 1 9 5 5 ~ ) Extraneuronal ; influences upon the electrical activity of

226

REFERENCES

isolated cortex following stimulation of the reticular activating system. Acta physiol. Scand. 33, 2-3, 169. Ivanov, I.I., Balabukha, V.S., Romantsev, E.F., and Fedorova, T.A., (1956); Metabolism and Radiation Sickness. Moscow. Iwama, K., and J a s p e r , H.H., (1957); The action of y-aminobutyric acid upon cortical electrical activity in the cat. J. Physiol., 138, 365. J a s p e r , H., (1949); Diffuse projection system: the integrative action of the thalamic reticular system. Electroenceph. clin. Neurophysiol., 1, 405. Jasper, H., Ricci, F., and Doan, B., (1958); Microelectrode analysis of cortical discharges during elicitation of conditioned reflexes in the monkey. Proc. int. Symp. on the EEG of higher nervous Activity, Moscow. Kaada, B.R., (1951); Somato-motor, autonomic and electrocorticographic responses to electrical stimulation of rhinencephalic and other s t r u c t u r e s in primates, c a t and dog. Acta physiol. Scand., 35, Suppl. 83. Kamiya, N., and Abe, S., (1950); Bioelectric phenomena i n the mycete plasmodium and their relation to protoplasmic flow. J. Colloid Sci., 5, 149. Kandel, E.R., Spencer, W.A., and Brinley, F.J., (1960); Transient and long-lasting electrical responses to direct hippocampal stimulation. Am. J. Physiol., 198, 687. Karamyan, A.I., (1956); Evolution of the Functions of the cerebral Cortex. Moscow. Karamyan, A.I., (1958); Influence of sympathic and adrenergic systems on the reflex activity on higher levels of the central nervous system. USSR J. Physiol., 44, 316. Karamyan, A.I., (1959a); Morphophysiological evolution of the higher p a r t s of the central nervous system. P r o c . 9th Congr. Physiologists, Biologists and Pharmacologists, 3, 89. Katz, B., and Schmitt, O.H., (1940); Electric interaction between two adjacent nerve fibres. J. Physiol., 97, 471. Kaufman, P.Y., (1912); Electrical phenomena in the c e r e b r a l cortex. Earliest Experiments on the EEG in Russia. Moscow, (1949), p. 139. Kedrov, A.A., and Naumenko, A.N., (1954); Problems of the Physiology of the intracranial Bloodflow and its clinical Implications. Moscow. Kempinsky, W.H., (1954); Steady potential gradients in experimental cerebral vascular occlusion. Electroenceph. clin. Neurophysiol., 6, 375. Keynes, R.D., (1951); The ionic movements during nervous activity. J. Physiol., 114, 119. Keynes, R.D., and Lewis, P.R., (1955); Electrolytes and function of the central nervous system. Neurochemistry. K.A.C. Elliott, I.H. Page and J.H. Quastel, Editors. Springfield, 440. Keynes, R.D., and Hodgkin, A.L., (1956); Some experiments with Ca on squid axons. XX Congr. int. Physiol., B r w e l l e s , 495. Kibyakov, A.V., (1933); On the humoral transmission of stimuli between neurones. Kazan med. J. 5/6, 457. Klosovsky, B.N., (1951); Blood Circulation in the Brain. Moscow. Knipet, I.N., (1955); Electrical activity of various levels of the c e r e b r a l cortex during the eliciting of defensive reflexes in the rabbit. Rep. Inst. of higher nervous Activity, 1, 294. Koechlin, B.A., (1954); Isothionic acid, the major anion in axoplasm of squid giant nerve fibres. Fed. Proc., 13, 80. Koelle, G.B., (1954); The histochemical localization of cholinesterase in the central nervous system of the rat. J. comp. Neurol., 100, 211. Kogan, A.B., (1949); Electrophysiological Research of the central Mechanism of certain complex Reflexes. Moscow. Koller, G., (1948); Rhythmische Bewegung und hormonale Steuerung bei den Malpigischen Gefibsen d e r Insekten. Biol. Zbl., 67, 201. Kolodnaya, A.Y., (1944); Bioelectrical phenomena in the c e r e b r a l cortex and subcortical structures during narcolan narcosis. Experiments in Biology and Medicine, 3, 31. Kondratyeva, P.P., (1958); Changes in the thermoregulation provoked by unifold

I WFE R EN C ES

227

general X-irradiation in rabbits dependent on the s t a t e of the hypothalamic a r e a . Med. Radiol. 1, 16. Kopylov, A.G., (1957a); Details of electroencephalographic Responses to rhythmic Stimulation by Light as an Indicator of the functional State of the human c e r e b r a l Cortex. Thesis, Leningrad University. Kopylov, A.G., (195713); Changes in the functional State of the Brain during Development of Sleep in Man as reflected in the EEG. Leningrad University, p. 15. Kornmuller, A.E., (1957); Zum Wesen d e s EEG auf Grund spezieller Untersuchungen. h t . Neurol. Congr., Rruxelles, 117. Korolkova, T.A., (1956); Research on physiological mechanisms of r e s p i r a t o r y rhythms in the electrical activity of the cortex. Rep. hst. of higher nervous Activity, 2, 181. Koshtoyants, K.S., (1951); Protein Metabolism and nervous Regulation. Moscow. Koshtoyants, K.S., (1957); Principles of comparative Physiology, 11. Moscow. Koshtoyants, K.S. , Byzov, A.L., and Mitropolitanskaya, R.D., (1954); Oscillographic r e s e a r c h of t h e nervous s y s t e m of the Bombyx m o r i in its different s t a g e s of development. J. Zool., 33, 807. Koshtoyants, K.S., and Kokina, N.N., (1957a); The r o l e of the acetylcholinecholinesterase s y s t e m in galvanotaxis and summation of stimuli in the Paramecium. Biophysics, 2, 46. Koshtoyants, K.S., and Kokina, N.N., (1957b); Rhythmic bioelectrical phenomena in a unicellular organism (Opalina r a n a r u m ) . Biophysics, 3, 422. Koshtyuk, P.G., (1959); The bineuronal Reflex Arc. ~ I O S C O W . Kozhevnikov, V.A., and Soroko, B.I., (1956); A double-cascade d.c. amplifier with high gain. Biophysics, 1, 95. Kozlova, A.V., (1959); Clinical application of radioactive isotopes for diagnostic and therapeutical purposes. P r o c . 2nd int. Conf. on the peaceful Application of atomic Energy, Section Radiobiology and Medicine, MOSCOW, p. 296. Krayevsky, N.A., (1957); Outline of Pathology of Radiation Sickness. Moscow. Krigel, E., and Nestianu, V., (1958); Electroencephalographic r e s e a r c h of cortical responses. J. of higher nervous Activity, 8, 570. Kristiansen, K., and Courtois, G., (1949); Rhythmic e l e c t r i c a l activity from isolated c e r e b r a l cortex. Electroenceph. clin. Neurophysiol., 1, 265. Krushinsky, L.V., (1954); Study of the relation between excitation and inhibition in the normal and pathological s t a t e by means of acoustic stimulation. P r o g r . contemp. Biol., 37, 74. Kryspin-Exner, W., (1943); Beitrage z u r Morphologie d e r Glia im Nissl-Bilde, Z. Anat. Entwickl. Gesch., 112, 389. Kuffler, S.W., and Vaughan-Williams, E.M., (1953); P r o p e r t i e s of slow skeletal muscle f i b r e s of frog, J. Physiol., 121, 318. Kurella, G.A., (1958a); A method for manufacturing intracellular electrodes. Biophysics, 3, 243. Kurella, G.A., (1958b); Methods of r e s e a r c h of dynamics of the steady s t a t e potential in various muscle t i s s u e s . Biophysics, 3, 614. Kuypers, C.G.M., (1956); Cortical projections t o the pons and the medulla oblongata in c a t and man. Anat. Rec., 124, 322. Kuzin, A.M., and Budilova, E.V., (1954); Change in the s t r u c t u r a l composition of nucleic acids of the b r a i n and spleen under influence of ionizing radiation. L e c t u r e s of the USSR Acad. Sci., 98, 961. Kvasov, D.G., (1948); On the cathodal depression of the nerve. Memorial Edition in Honour of A.V. Leontovitch, p, 238. Lapitzkii, D.A., (1948); An Example of functional Analysis of s o m e pathological P r o c e s s e s . Leningrad. Larionov, V.E., (1899); Galvanometric measurement of the c u r r e n t s of the c e r e b r a l cortex in normal and pathological s t a t e s by acoustic stimulation. P r o g r . contemp. Biol., Moscow, 92, 1949. Latash, L.P., (1961); Electroencephalographic changes after administration of

228

REFERENCES

adrenalin in small doses in healthy people and patients with a lesion of the hypothalamus. Korsakov J. Psychopathol. Psychiat., 61, 1169. Lazarev, P.P., (1916); Studies on the ionic Theory of Excitability. Moscow. Lego, A.A.P., (1944a); Spreading depression of activity in the cerebral cortex. J. Neurophysiol., 7, 339. Letlo, A.A.P., (194413); Pial circulation and spreading depression in the cerebral cortex. J. Neurophysiol., 7, 391. LeZo, A.A.P. , (1951); The slow voltage variation of cortical spreading depression of activity. Electroenceph. clin. Neurophysiol., 3, 3, 315. Lego, A.A.P., and Morison, R.S., (1945); Propagation of spreading cortical depression. 2. Neurophysiol., 8, 33. Lebedinsky, A.V., (1926); Influence of sympathic innervation on the electrical conductivity of striated muscle. USSR J. Physiol., 9, 183. Lebedinsky, A.V., (1933); Contribution t o the analysis of sympathic influence on striated muscle tissue. USSR J. Physiol., 16, 111. Lebedinsky, A.V., (1957a); Some neuro-endocrine aspects of the response of the organism to ionizing radiation. Med. Radiol., 2, 35. Lebedinsky, A.V., (1957b); On the sources and mechanisms of the effect of small doses of ionizing radiation on the organism. Proc. nat. Conf. on the Use of radioactive and stabile Isotopes in Economy and Science, Moscow. Lebedinsky, A.V., Grigorev, F.G., and Demirchoglyan, G.G., (1959); On the biological action mechanism of ionizing radiation in small doses. Proc. int. Conf. on the peaceful Use of atomic Energy, Section Radiobiology and Medicine, 5. Lehmann, J.E., (1937); The effect of changes in pH on the action of mammalian nerve fibers. Amer. J. Physiol., 118, 600. Li, C.L. , (1959); Cortical intracellular potentials and their responses to strychnine. J. Neurophysiol., 22, 4, 444. Li, C.L., Cullen, C., and Jasper, H., (1956a); Laminar microelectrode studies of specific somatosensory cortical potentials. J. Neurophysiol., 19, 111. Li, C.L., Cullen, C., aqd Jasper, H.H., (1956b); Laminar microelectrode analysis of cortical unspecific recruiting responses and spontaneous rhythms. J. Neurophysiol., 19, 131. Li, C.L., and Jasper, H., (1953); Micro-electrode studies of the electrical activity of the cerebral cortex in the cat. J. Physiol., 121, 117. Liberman, E.A., (1958a); Methodology of research of the effect of ionizing radiation on the function of nerve cells. Biophysics, 3, 241. Liberman, E.A. , (195813); Dosimetry of radioactive Isotopes. Moscow. Libet, B., and Gerard, R.W., (1939); Effects of ions on brain activity. J. Neurophysiol., 2, 153. Libet, B., and Gerard, R.W., (1941); Steady potential fields and neurone activity. J. Neurophysiql., 4, 438. Ling, G., (1957); Fixed charge hypothesis for a key mechanism in change of Na-K pwmeability during excitation. Fed. Proc., 16, 81. L i s s a , K., (1955); New experimental aspects in the study of diencephalic mechanisms and processes of higher nervous activity. J. of higher nervous Activity, 5, 636. Liss&, K., Endorbczi, E., and Hasznos, T., (1952); Effect of cortical denervation upon acetylcholin-cholinesterase system and excitability of the central nervous system. Acta Physiol. Academ. occentiarum hungaricae, 8, Fasc. 1, 39. LivGov, M.N., (1944a); Electrical activity of the cerebral cortex in animals and man in normal and pathological states. Lectures of the USSR Acad. Sci., Biol. Series, 6, 319. Livanov, M.N., (1944b); Electrical activity of the cerebral cortex in man and animals in normal and pathological states. Bull, USSR Acad. Sci., 6, 331. Livanov, M.W., (1958a); Central and peripheral interaction in radiatien sickness. Me& Radiol., 1, 19. Livanov, M.N. , (1956b); Changes occurring in various sections of the central nervous system in animals after X-irradiation. Proc. nat. Conf. on medical Radiology, 3.

REFERENCES 229 Livanov, M.N., (1959); Electrophysiological studies of radiation injuries. Proc. 9th Congr. of Physiologists, Biologists and Pharmacologists, 1, 275, Livingston, R.B., (1957); Neurophysiology of reticular formation. Symp. Brain Mechanisms and Drug Action. Oxford, p. 3. Longo, V.G., Berger, G.P. Von, and Bovet, D., (1954); Action of nicotine and of the "ganglioplegiques centaux" on the electrical activity of the brain. J. Pharmacol. exp. Therap., 111, 3, 349. Longo, V.G., and Silvestrini, B., (1957); Effect of adrenergic cholinergic drugs injected by intracarotid route on electrical activity of brain. P r o c . SOC.exp. Biol., 95, 43. Lorente de N6, R., (1939); Transmission of impulses through cranial motor nuclei. J. Neurophysiol., 2, 402. Lumsden, C.E., (1957); The problem of correlation of quantitative methods and tissue morphology in the central nervous system. D. Richter, Editor. Metabolism of the Nervous System. London, Pergamon Press (pp. 91-100). Lumsden, C.E., and Pomerat, C.M., (1951); Normal oligodendrocytes in tissue culture. Exp. Cell. Res., 2, 103. Lundberg, A., (1955); The electrophysiology of the submaxiliary gland of the cat. Acta physiol. Scand., 35, 1. Lyev, A.A., (1952); The Method of Evaluation of the functional State of the human and animal cerebral Cortex by EEG Reactivity Curves. Thesis, Moscow. Lyubimov, N.N., (1958); Electrical changes in the cortical and hypothalamic structures in the propagation of conditioned reflexes. J. of higher nervous Activity, 8, 560. MacIntosh, F., and Oborin, P., (1953); Release of acetylcholine from intact cerebral cortex. Abstr. Comm., XtX Int. Physiol. Congr., Montreal, p. 580. Magoun, H.W., (1950); Caudal and cephalic influences of the brain stem reticular formation. P h y s i d . Rev., 30, 459. Magoun, H.W.,(1952); An ascending reticular activating system in the brain stem. Arch. Neurol. Psychiat., 67, 145. Magoun, H.W., (1958); The waking Brain. Springfield, Thomas. Maiorov, F.P., Pavlov, I.P., and Lipatova, N., (1956);-Changes in the higher nervous activity in dogs evoked by X-irradiation of the cervical part of the autonomic nervous system. Rep. Pavlov Inst. Physiol., 5, 79. Mantegazzini, K., Poek, G., and Santibanez, H., (1959); The action of adrenaline and nor-adrenaline on the cortical electrical activity of the "encbphale isol6" of cat. Arch. ital. Biol., 97, 222. Mwqkhin, B.N., and Buznikov, G.A., (1960); A new biological method for the quantitative analysis of serotonine. USSR J. Physiol., 46, 1180. Marazzi, A.S., and Lorente de N6, R., (1944); Interaction of neighbouring fibers in myelinated nerve. J. Neurophysiol., 7, 83. Marazzi, AS., '(1953); Some indications of cerebral humoral mechanisms, Science, 118, 367. Mashkovsky, M.D., Liberman, S.S., and Folejayeva, A.I., (1955); On the Ph~macolo€3 of m i n a z i n e . Pharmacol. Toxicol., 18, 14. Maslova, M.N., (1959); R e s e u c h on the effect of cholinolytic and anticholinesterase writs on the c e r e b r a l cortex and the reticular fbrmation of the brain stem. Proc. 9th Congr. Physiologists, Biologists and Pharmacologists, Moscow-Minsk, 2, 173. McCann, S;M., (1957); The ACTH-releasing activity in vivo. Endocrinology, 60, 664. McCann, S.M., and Brobeck, J.R., (1954); Evidence for a role of the supraopticohypophyseal system in regulation of adrenocorticotrophin secretion. Proc. SOC. exp. Biol. Med., 87, 318. McCann, S.M., and Sydor, K.L., (1954); Blood and pituitary adrenocorticotrophin in adrenalectomized rats with hypothalamic lesions. Proc. SOC.exptl. Biol. Med., 87, 369. Mcnwain, H., (1951); Metabolic response in vitro to electrical stimulation of sections of mammalian brain. Biochem. J., 49, 382. Mcnwain, H., 1957); Chemotherapy and the Central Nervous System. London, McIlwain, H., 1959); Biochemistry and the Central Nervous System. London.

t

230

REFERENCES

McLardy, T., (1956); Some midbrain-thalamic longfiber connections in monkey. P r o g r e s s in Neurobiology. P r o c . f i r s t int. Meeting Neurobiologists, Amsterdam, p. 282. McLennan, H., (1957); Diffusion of K, inulin and thiocyanate in the extracellular spaces of mammalian muscle. J. Physiol., 139, 79. Mescherskii, R.M., (1958); The influence of local irradiation of the cortical endings of the visual analyser in the rabbit on the defensive motor reflexes elicited by light stimuli. Publ. Inst. higher nervous Activity, 4, 47. Mescherskii, R.M., (1959); A stereotaxic device for s m a l l laboratory animals. USSR J. Physiol., 45, 498. Miahle-Voloss, C., and Benoit, J., (1954); L'intermedine dans l'hypophyse et l'hypothalamus du canard. C.R. SOC.Biol. Paris, 148, 56. Mikhelson, M.Y., (1957); The physiological Role of Acetylcholine and the Search for new therapeutic Drugs. Leningrad, p. 9. Mikhelson, M.Y., (1959); The mechanism of action of acetylcholine and the s e a r c h for new therapeutic drugs. 9th Conf. Physiologists, Biochemists and Pharmacologists, Moscow, 3, 211. Mikhelson, S.Y., (1948); The basic properties of cholinesterase. Progr. contemp. Biol., 25, 321. Minut-Sorokhtina, O.P., (1934); The influence of the brain nuclei on the spinal reflexes in the frog. USSR J. Physiol., 17, 718. Moiseyev, E.A., (1957); On the pathogenesis of endocrine disorders in relation to radiation injuries. P r o c . Conf. Use of radioactive and stabile Isotopes in Economy and Science, p. 24. Monnier, A.M., (1952); P r o p e r t i e s of nerve axons (11). The damping factor as a functional criterion in nerve physiology. Cold Spr. Harb. Symp. quant. Biol., 17. Morison, R.S., and Dempsey, E.W., (1943); Mechanism of thalamo-cortical augmentation and repetition. h e r . J. Physiol., 138, 297. Morrell, F., (1960); Microelectrodes and steady potential study suggesting a dendritic locus in closure. The Moscow Colloquium on EEG of higher nervous Activity. Electroenceph. clin. Neurophysiol., Suppl. 13, 65. Moruzzi, G., Broockhart, J.M., Niemer, W.T., and Magoun, H.W., (1950); Augmentation of evoked electrocortical activity during spindle bursts. Electroenceph. clin. Neurophysiol., 2, 29. Moskalenko, Y.E., and Naumenko, A.N., (1957); The fluctuations in the movement of the cerebrospinal fluid in spinal cord and brain of animals. USSR J. Physiol., 43, 928. Murphy, I.P., and Gellhorn, E., (1945); The influence of hypothalamic stimulation on cortically induced movements and on action potentials of the cortex. J. Neurophysiol., 8, 341. Naitoh, I., (1958); Direct current stimulation of Opalina with intracellular microelectrode. Anat. Zool. Japan., 31, 1, 59. Narikashvili, S.P., (1953); The influence of stimula of subcortical centres on the electrical activity of the c e r e b r a l cortex. Rep. Inst. Physiol. Georgian Acad. Sci., 9, 133. Narikashvili, S.P., (1958); On the influence of the thalamic reticular system on the activity of specific sensory elements in the cerebral cortex. Proc. 8th Conf. higher nervous Activity, Leningrad. Narikashvili, S.P,, (1959); Some problems of the physiology of the unspecific structures of the brain. P r o c . 9th Conf. Physiologists, Biochemists and Pharmacologists, 3, 37. Narikashvili, S.P., and Moniava, E.S., (1957); On the interaction of diffuse and specific thalamo-cortical projection systems. Rep. Georgian Acad. Sci., 19, 347. Nathan, P., and Aprison, M.H., (1955); Cholinesterase activity in cytoplasmic particles from rabbit brain. Fed. Proc., 14, 106. Naumova, T.S., (1956); The so-called arousal s y s t e m s of the brain stem. Korsakov J. Neuropathol. Psychiat., 56, 668. Nauta, W.J.H., (1956); An experimental study of the fornix system i n the rat. J . comp. Neurol., 104, 247.

REFERENCES

231

'Nauta, W.J.H., (1958); Hippocampal projection and related neural pathways to the midbrain in the cat. Brain, 81, 30. Niemer, W.T., Powell, E.W., and Goodfellow, E.F., (1960); The subcortex and hypothalamic afterdischarge in the cat. Electroenceph. clin. Neurophysiol., 12, 345. Novikova, L.A., and Farber, D.A., (1956); Electrophysiological r e s e a r c h on the connection between the visual and the auditory analyser under the influence of a dominating focus in the cerebral cortex of the rabbit. USSR J. Physiol., 42, 341. Novikova, L.A., Pusinov, V.S., and Semiokhina, A.F., (1952); Electrophysiological analysis of the occlusive function of the cerebral cortex in the rabbit under the influence of a dbminating focus. J. higher nervous Activity, 2, 334. Ochs, S., (1956); The direct cortical response. J. Neurophysiol., 19, 6 , 513. Orbeli, L.A., (1923); The mechanism of the ontogenesis of cerebrospinal connections. Rep. Lesgaft Inst., 6, 202. Orbeli, L.A., (1935); Readings on the Physiology of the nervous System. Leningrad. Orbeli, L.A., (1955); The effect of ionizing radiation'on the animal's organism. Proc. USSR Acad. Sci. on the peaceful Use of atomic Energy. Orbeli, L.A., and Tonkikh, A.V., (1928); The role of the sympathic nervous system in raising the temperature during heat stroke. Proc. 3rd nat. Conf. Physiologists, p. 243. Osterhout, W.I.V., (1922); Injury, Recovery and Death. Philadelphia. Palade, G.E., and Palay, S.L., (1954); Electron microscope observations of interneuronal and neuromuscular synapses. Anat. Rec., 118, 335. Palay, S.L., and Palade, G.E., (1955); The fine structure of neurons. J. biophys. biochem. Cytol., 1, 69. Palladin, A.V., (1953); Metabolism in the brain during inhibition and excitation of higber nervous activity. Proc. 19th int. Congr. Physiologists, Moscow, p. 129. Palladin, A.V., (1955); Biochemical processes in the brain in different functional conditions. Proc. 8th Conf. Physiologists, Biochemists and Pharmacologists, Moscow, p. 465. Papez, I.M., (1937); A proposed mechanism of emotion. Arch. Neurol. Psychiat., 38, 725. Pavlov, I.P., (1932); The Physiology of higher nervous Activity. Reprinted in: Complete works. Moscow-Leningrad, 3, (1949), 480. Pavlov, I.P., (1949); The Etiology of Neuroses. Reprinted in: "Pavlovskiye sredy", 1, 268. Pavlov, I.P., (1949a); How the Anodonta opens its Shells. Complete works. MoscowLeningrad ( F i r s t pilblished 1885). Pavlov, I.P., (1951); An EssAi on the physiological Conception of the Symptomatology of Hysteria. Complete works 3, 210. Pavlygina, R.A., (1959); Destruction and recovery of higher nervous activity during existence of a dominating focus in the hypothalamic a r e a of the rabbit brain. Rep. Inst. higher nervous Activity; Physiological Series, 3, 19. Pecher, C., (1939); L a fluctuation d'excitabilitg de la fibre nerveuse. Arch. int. Physiol., 49, 129. Penfield, W., (1924); Oligodendroglia and its relation t o classical neuroglia. Brain, 47. 430. P e k i e l d , W., (1930); A further modification of Del Rio Hortega's method of staining oligodendroglia. Amer. J. Pathol., 6, 445. Penfield, W., and Jasper, H., (1958); Epilepsy and functional Anatomy of the human Brain. Little Brown, Boston. Pernow, B., (1953); Studies on substance P. Acta physiol. Scand., 29, Suppl. 105. P e r r y , W.L.M., (1953); Acetylcholine release in the cat's superior cervical ganglion. J. Physiol., 119, 439. Phillips, C.G., (1955); The dimensions of a cortical motor point. J. Physiol., 129, 20. Phillips, C.G., (1956a); Intracellular records from Betz cells in the cat. Quart. J. exp. Physiol., 41 , 58.

232

REFERENCES

Phillips, C.G., (1956b); Cortical motor threshold and the thresholds and distribution of excited Betz cells in the cat. Quart. J. exp. Physiol., 41, 70. Phylippson, M., (1930); Sur la r6sistance 6lectrique du muscle mesuree 5 hautes frkquences. C.R. SOC.biol., 83, 1399. Pickford, M., (1939); The inhibitory effect of acetylcholine on water diuresis in the dog, and its pituitary transmission. J. Physiol., 95, 226. Pope, A., (1952); Quantitative distribution of dipeptidase and acetylcholine esterase in architectonic l ayer s of rat cerebral cortex. J. Neurophysiol., 15, 115. P o r te r , K.R., (1953); Observations on submicroscopic basophilic component of cytoplasm. J. exptl. Med., 97, 727. P o r te r , K.R., (1954); Electron microscopy of basophilic components of cytoplasm. J. Histochem. Cytochem., 2, 346. P o r te r , R.W., (1952); Alteration in electrical activity of the hypothalamus induced by s t r e s s stimulus. Amer. J. Physiol., 169, 629. Powell, T.P.S., and Cowan, W.M., (1956); A study of thalamo-striate relations in the monkey. Brain, 79, 364. Prawdicz-Neminski, W.W., (1913); Ein Versuch der Registrierung der elektrischen Gehirnerscheinungen. Zbl. Physiol., 27, 951. Prawdicz-Neminski, W.W., (1925); Zur Kenntnis der elektrischen und der Innervationsvorgkge in den funktionellen Elementen und Geweben d e s T i ers. Pfl. Arch., 209, 362. P r o s s er , C.L., (1934a); Effect of the central nervous system on responses of light in Eisenic Fetida, SAV. J. comp. Neurology, 59, 1, 61. P r o s s er , C.L., (1934b); Action potentials in the nervous system of the crayfish. I. Spontaneous impulses. J. cell. comp. Physiol., 4, 185. P r o s s er , C.L., (1943); Ions on activity of crayfish abdominal ganglia. J. cell. comp. Physiol., 22, 131. Purpura, D.P., (1956); A neurohumoral mechanism of reticulocortical activation. Amer. J. Physiol., 186, 250. Purpura, D.P., (1957); Mechanism of the inhibitory action of LSD on cortical dendritic activity. Ann. N.Y. Acad. Sci., 66, 3. Purpura, D.P., Girado, M., and Grundfest, H., (1957a); Mode of action of aliphatic amino acids on cortical synaptic activity. Proc. SOC.exp. Biol. Med., 95, 791. Purpura, D.P., Girado, M., and Grundfest, H., (1957b); Selective blockade of excitatory synapses in the cat brain by y-aminobutyric acid (GABA). Science, 125, 1200. Purpura, D.P., and Grundfest, H.,(1956); Nature of dendritic potentials and synaptic mechanisms in cerebral cortex of cat. J. Neurophysiol., 19, 6, 573. Purpura, D.P., and Grundfest, H., (1957); Physiological and pharmacological consequences of different synaptic organizations in cerebral and cerebellar cortex of cat. J. Neurophysiol., 20, 5. Rabinovich, M.Y., (1958); Electrical activity of particular cortical layers of the motor and auditory analysers during propagation of internal inhibition. J. higher nervous Activity, 9, 107. Ravitz, L.I., (1951); Standing potential correlated of hypnosis and narcosis. Arch. Neurol. Psychiat., 65, 4, 413. Ravitz, L.I., (1953); Electrodynamic field theory in psychiatry. Sth. Med. J., 46, 7, 650. Roberts, E., and Frenkel, S., (1950); y-Aminobutyric acid in brain. Fed. Proc., 9, 1, 219. Robertson, J.D., (1956); The ultrastructure of a reptilian myoneuronal junction. J. biophys. biochem. Cytol., 2, 1, 318. Robiner, I.S., (1954); Electrical activity of the cortex and thalamus of the cat under ether narcosis. USSR J. Physiol., 40, 404. Roeder, K.D.,(1941); The effect of potassium and calcium on the spontaneous activity of the isolated crayfish nerve cord. J. cell. comp. Physiol., 18, 1.

REFERENCES

233

Roitbak, A.I., (1950); Electrical potentials generated in the cerebral cortex by d.c. stimulation. Proc. Conf. Georgian SOC.of Physiologists. Roitbak, A.I., (1953a); Bioelectrical phenomena in the cerebral cortex generated by peripheral stimulation. Proc. 13th Session med. biol. Section Georgian Acad. Sci., Tbilisi, p. 28. Roitbak, A.I., (1953b); On the spreading of impulses from the respiratory centre to the cortex in vertebrates. Lectures of the Georgian Acad. Sci., 14, 361. Roitbak, A.I., (1955); Bioelectrical Phenomena in the cerebral Cortex. Tbilisi. Rojansky, N.A., (1956); Evolutionary principles of the development of the cerebral cortex. Proc. C o d . on the physiological Evolution of the nervous System, p. 138. Rossi, J.F., and Zanchetti, A., (1960); The Reticular Formation of the Brain Stem. MOSCOW, p. 147. Rothenberg, M.A., (1950); Studies of permeability in relation to nerve function. n. Ionic movements across axonal membranes. Biochim. biophys. Acta, 4, 96. Rothballer, A.B., (1956); Studies on the adrenalinesensitive component of the reticular activating system. Electroenceph. clin. Neurophysiol., 8, 603. Rusinov, V.S., (1947); The electrophysiological method as applied to the theory of integration activity of the nervous system. Proc. 7th Conf. Physiologists, Biochemists and Pharmacologists, Moscow, p. 201. Rusinov, V.S., (1953); The electrophysiological analysis of the occlusive function of the cerebral cortex during existence of a dominating focus. Proc. int. Conf. of Physiologists, Moscow, p. 147. Rusinov, V.S., (1954); Rhythms of bioelectrical potentials in the cerebral cortex as the projection of essential nervous processes. Nature and Research of bioelectric Potentials, p. 57. Rusinov, V.S., (1955); The theories of Vvedensky and Ukhtomsky on inhibition and its relation t o the theories of I.P. Pavlov. J. higher nervous Activity, 5, 305. Rusinov, V.S., (1958a); General and local changes in the EEG during elicition of a conditioned reflex in the human central nervous system. Proc. int. Colloquium on the EEG of higher nervous Activity, Moscow. Rusinov, V.S., (1958b); Electrophysiological research of foci of steady state impulses in the central nervous system. J. higher nervous Activity, 8, 473. Rusinov, V.S., and Rabinovitch, M.Y., (1958); Electroencephalographic researches in the laboratories and clinics of the Soviet Union. Electroenceph. clin. Neurophysiol., Suppl. 8, 3. Samoilov, A.F., (1925); On the transmission of impulses from the motor nerve to the muscle. Memorial edition in hpnour of the 75th anniversary of I. Pavlov, p. 75. Sapegno, E., (1930); Uber die Impedanz und KapazitXt des quergestreiften Muskels in L2ngs-und Querrichtung. Pfl. Arch., 224, 187. Sarkisov, S.A., (1935); Bioelectric phenomena in the cerebral cortex under the influence of narcotic drugs. USSR J. Neuropathol. Psychiat. psychic Health, 4, 1. Sarkisov, S.A., (1937); On the relation of bioelectrical potentials to the anatomic areas and their boundaries. J. Neuropathol. Psychiat., 6, 139. Sarkisov, S.A., (1956); The relation between structure and function of the cerebrum. Proc. 20th int. Congr. of Physiologists, Brussels. Sarkisov, S.A., and Livanov, M.N.,(1933); The bioelectric phenomena and their localisation in the cerebral cortex. Conf. Neuropathol., Psychiat. psychic Health, 2,l. Sarkisov, S.A., and Polyakov, G.I., (1949); Cytoarchitectonics of the cerebral cortex in man. Moscow. Sawyer, C.H., Everett, J.W., and Green, J.D., (1954); Rabbit Horsley-Clarke coordinates. J. comp. Neurol., 101, 801. Scanavi, G.I., (1948); Physics of dielectric Media. Moscow. ScharFer, B., (1941); Neurosecretion. 11. Neurosecretory cells in the c e n t r d nervous system of cockroaches. J. comp. Neurol., 74, 93. Scharrer, B., and Scharrer, E., (1944); Neurosecretion. VI. A comparison between the intercerebralis-cardiacum-allatum system of the insects and the hypothalmohypophyseal system of the certebrates. Biol. Bull., 87, 242.

234

REFERENCES

Scharrer, E., and Scharrer, B., (1945); Neurosecretion review. Physiol. Rev., 25, 171. Scheibel, M., and Scheibel, A., (1958); Neurons and neuroglia cells as seen with the light microscope. Symp. Biology of neuroglia. Springfield, p. 5. Scheminzky, F., Hochsttidt, O.,and Adler, P., (1936); Uber das Wesen der Galvanonarkose beim Frosch. Pfl. Arch., 237, 284. Scott, B.J.H., (1957); Electric oscillations generated by plant roots and a possible feedback mechanism responsible for them. Aust. J. biol. Sci., 10, 2, 164. Scott, B.J.H., McAuley, A.L., and Jeves, P., (1955); Correlation between the electric current generated by a bean root growing in water and the rat e of elongation of the root. Aust. J. biol. Sci., 8, 1, 36. Sechenov, I.M., (1882); Galvanic phenomena in the medulla oblongata of the frog. Earliest Experiments on the EEG in Russia,(1949), p. 13. Sepp, E.K., (1928); Die Dynamik der Blutzirkulation im Gehirn. Berlin. Shevelyova, V.S., (1945); Mechanism of stimulus transmission in the superior cervical ganglion. USSR J. Physiol., 31, 157. Shevelyova, V.S., (1953); Interneuronal synaptic Transmission of nerve Impulses. Thesis, Leningrad University. Shevelyova, V.S., (1956); The significance of calcium and potassium ions in the synaptic transmission of nerve impulses. Lectures of the USSR Acad. Sci., 109, 882. Sholl, D.A., (1956); The organization of the Cerebral Cortex. London. Shvets, T.B., (1958a); Slow electrical processes in the cerebral cortex of the rabbit. Proc. Conf. on the Electrophysiology of the central nervous System, Moscow, 136. Shvets, T.B., (1958b); Comparison of the EEG observed during existence of a dominant focus, with the EEG in the generalisation stage of a conditioned reflex. Proc. 18th Conf. on higher nervous Activity, 3, 193. Sloan, N., and Jasper, H., (1950); The identity of spreading depression and suppression. Electroenceph. clin. Neurophysiol., 2, 59. Smirnov, G.D., (1953); The lability of neural processes in the central and peripheral parts of the visual analyser. J. higher nervous Activity, 3, 941. Smirnov, G.D., and Aladjalova, N.A., (1956); Electrical activity and impedance of the cerebral cortex. Lectures of the USSR Acad. Sci., 106, 573. Smirnov, V.E., (1950); Influence of galvanic current on the condition of human muscle tissue. Mechanisms of pathological Reactions, 16, 381. Snesarev, P., (1926); Uber einen degenerathen Prozess im Hirn im Zusammenhang mit seiner klinischen Manifestation. Degenerative Glia und Kornelungen des Hirns. Zschr. ges. Neurol. Psychiat., 106, 702. Snezhko, A.D., (1958); Vital observation on the degeneration of oxidizing processes in the central nervous system. Proc. nat. Conf. on the Application of radioactive and stabile Isotopes in Agriculture and Science. Section Radiobidogy, p. 177. Sokolova, A.A., and Khon Sek Bu, (1957); Electrophysiological research of a dominating focus in the cerebral cortex of the rabbit elicited by d.c. stimulation. J. higher nervous Activity, 7, 135. Stutinsky, F., (1953); Modalit& de la neuros6cretion dans le noyau prCoptique de l'anguille. Bull. SOC.Zool. France, 77, 240. Sverdlov, S.M., (1941); Electrophysiological observations on nerves and muscles of the Anodonta. Rep. Kazan State Univ., 101, 4. Sverdlov, S.M., (1943); Electrophysiological observations on nervous tissue of Anodonta cygnea. Rep. Kazan State Univ., 11, 31. Sverdlov, S.M.,(1954); Electrical potentials on the ganglion of Anodonta. Rep. Armenian Acad. Sci., 7, 69. Tarusov, B.N., (1935); The dielectric constants of muscle tissue. Research of the physical Chemistry of the Cell, p. 183. Tarusov, B.N., (1939a); The Kinetics of the pri m ary arousal Reaction. Thesis, Moscow.

RE FE RE NCE S

235

Tarusov, B.N., (1939b); Electrical depolarisation of the living cell. Problems of Permeability. Proc. Conf. of Physiologists, Biochemists and Pharmacologists, Moscow. Tasaki, I., Polley, E.H., and Orrego, F . , (1954); Action potentials from individual elements in cat geniculate and striate cortex. J. Neurophysiol., 17, 454. Tauc, L., (1954); RBponse de la cellule nerveuse du ganglion abdominal d'Aplysia depilans ii la stimulation directe intracellulaire. C. R. Acad. Sci. Paris, 339, 1537. Tauc, L., (1955); RCponse de la cellule nerveuse du ganglion abdominal d'Aplysia punctata activCe par voie synaptique. J. Physiol. Paris, 47, 286. Teorell, T., (1946); Application of square wave analysis to bioelectric studies. Acta physiol. Spnd., 12, 235. Teorell, T., (1958); Transport processes in membranes in relation to the nerve mechanism. Exp. cell. Res., Suppl. 5, 83. Thompson, R.H.S., and Thickner, A., (1953); Cholinesterase activity of arteries. J. Physiol., 121, 623. Tobias, I., (1955); Effects of phospholipases, collagenase and chymotrypsin on impulse conduction and resting potential in the lobster axon with parallel experiments on f r o g muscle. J. cell. comp. Physiol., 46, 183. Tonkikh, A.V., (1957); The role of the adrenal gland in the development of radiation sickness. P r o c . Conf. on the Use of radioactive and stabile Isotopes in Agriculture and Science, Moscow, p. 25. Tonkikh, A.V., (1959); On the relation between the hypothalamus and the ductless glands. Proc. 9th Congr. Physiologists, Biochemists and Pharmacologists, 3, 143. Torda, C., (1953); Effect of convulsion inducting agents in the acetylcholine content and on the electric activity of the brain. Amer. J. Physiol., 173, 179. Tower, D.B., (1954); Structural and functional organization of mammalian cerebral cortex: correlation of neurone density with brain size; cortical neurone density in finwhale (Balaenoptera physalus L.) with note on cortical neurone density in Indian elephant. J. comp. Neurol., 101, 19. Trivus, S.A., (1900); Action.potentials in the cerebral cortex elicited by peripheral stimulation. Earliest Experiments on the EEG in Russia, (1949), p. 111. Trofimov, L.G., Lyubimov, N.N., and Naumova, T.S.,(1958); Relations between the cortex and the subcortex in the propagation of feeding and defensive conditioned reflexes in the dog. Proc. int. Colloquium on the EEG of higher nervous Activity, Moscow. Tschirgi, R.D., and Taylor, J.L., (1958); Slowly changing bioelectric potentials associated with the blood-brain b a r r i e r . Amer. J. Physiol., 195, 7. Tsypin, A.B., (1956); The effects of ionizing radiation on the condition of the visual analyser of the rabbit. Med. Radiol., 1, 21. Twarog, B.M., and Page, T.H., (1953); Serotonin content of some mammalian' tissues and urine and a method f o r its determination. Amer. J. Physiol., 175, 157. UteV*ii, A.M., (1959); The role of hormonal metabolism and i t s mechanism. Wechanisms of the Action of Hormons. Kiev, p. 27. Van Harreveld, A., (1957); Asphyxial volume .hanges of cortical cells and fibers. Fed. Proc., 16, 131. V ~ J Harreveld, I A., and Ochs, S., (1957); Electrical and vascular concomitants of Spreading depression. Amer. J. Physiol., 189, 159. Van Harreveld, A., and Stamm, IS., (1953); Vascular concomitants of the spreading cortical depression. J. Neurophysiol., 150, 487. Vassilevskii, V.M., and Naumenko, A.N., (1959); Blood Velocity in the Brain and the Movement of the cerebrospinal Fluid. MOSCOW. Vassilyev, L., (1939); Electrotonic recovery of physiological functions. Publ. of the Brain Inst., 7. Vassilyeva, V.M., (1955); Changes in the electrical Activity of the Brain in Rats during reflexogenic epileptic Seizures. Thesis, Moscow. Vassilyeva, V.M., (1958); Electroecephalographic r e s e a r c h on the motor a r e a of

236

REFERENCES

the cerebral cortex in white rats during epileptic seizures. J. higher nervous Activity, 8, 602. Verigo, B.F., (1889); Action potentials in the frog brain. Earliest Experiments on EEG in Russia, (1949), p. 49. Verigo, B.F., (1901); Cathodal depression. Its background and significance for the EEG. Physiology of the nervous System. Moscow, p. 548. Vinogradov, M.I., (1952); The Theory of N.E. Vvedensky on fundamental nervous Processes. Moscow. Vogt, M., (1954);The concentration of sympathin in different parts of the central nervous system under normal conditions and after the administration of drugs. J. Physiol., 123, 451. Vogt, M., (1959);Adrenalin and noradrenaline in nervous tissue. Biochemistry of the Central Nervous System. Proc. of the IV int. Congr. Biochemistry, Vienna., p. 5, Pergamon Press. Vogt, M.,(1959a); Catecholamines in brain. Symp. on Catecholamines, Baltimore, Williams and Wilkins (p. 487). Volodina, A.V., (1958); -Influence of drugs on stimulus propagation in specific and diffuse afferent systems. New Data on the Pharmacology of the Reticular Formation and the Synaptic Transmission. Leningrad, p. 13. Von Euler, C., (1950); Slow temperature potentials in the hypothalamus. J. cell. comp. Physiol., 36, 333. Von Euler, C., (1953); A preliminary note on slow hypothalamic "osmopotentials". Acta physiol. Scand., 29, 133. Von Euler, U.S., and Gaddum, J.H., (1931);An unindentified depressor substance in certain tissue extracts. J. Physiol., 72, 74. Vvedenskii, N.E., (1884);"Telephonic" research in electrical phenomena in muscle and neural apparatus. Sel. Works, 1, 1950. Wagner, R., and ,Mitchell, G., (1958); Genetics and Metabolism. Moscow. Wagner, W., (1913);Zur Theorie der unvollkommenen Dielectrica. Ann. Phys.,

40, 18.

Ward, A.A., and McCulloch, W.S., (1947); Projection of frontal lobe on hypothalamus. J. Neurophysiol., 10, 309. Whitlock, D.G., Arduini, A., and Moruzzi, G., (1953); Microelectrode analysis of pyramidal system during transition from sleep to wakefulness. J. Neurophysiol.,

16, 414.

Whitlock, D.G., and Nauta, W.J., (1956);Subcortical projections from the temporal neocortex in macaca mulatta. J. comp. Neurol., 106, 183. Wilson, L.D., Weinberg, J.A., and Bern, H.A., (1957); The hypothalamic neurosecretory system of the tree frog Hylla Regilla. J. comp. Neurol., 107, 2, 253. Woronzow, D.S., (1924); Wie schnell steigt der konstante Strom die Leishngsfiihigkeit der mit einigen Salzen behandelten Nerven wieder an. Arch. ges. Physiol., 207, 279. Wyckoff, R.W.G., and Young, J.Z., (1956); The motoneurone surface. Proc. roy. SOC.

B., 144, 440.

Yurman, N., (1900);Anatomical and physiological Research of the Substantia Nigra. Thesis, University of St. Petersburg. Zakutinsky, D.I., (1956);Late Effects of Injuries caused by ionizing Radiation. Moscow, p. 3. Zarubashvili, A.D., and Naneishvili, B.P., (1956); Contributions to pathological Problems of Radiation Injuries. Tbi1isi.Zavarzin, A.A., and Schelkunov, S.I., (1954); A Guide to Histology. Leningrad. Zhukov, E.K., (1946a); The excitation waves in the nerve of vertebrates. Progr. contemp. Biol., 22, 23. Zhukov, E.K., (1946b); The evolution of qxcitability. Publ. in Honour of the Anniversary of the Leningrad University, Section Biology, 9. Some laws of evolution of excitability. J. gen. Biol., 7, 435. Zhukov, E.K., (1946~); Zhukov, E.K., (1947); Local excitability. Pregr. contemp. Biol., 23, 23.

REFERENCES

237

Zhukov, E.K., (1948); Nonlinear laws of activity of excitable systems. Lectures of the Leningrad University, 8, 29. Zhukova, G.P., (1958); Some aspects of the neuronal structure of the reticular formation of the brain stem. Proc. Conf. of the Moscow Brain Inst. on Structure and Function of the Reticular Formation and its Place in the System of Analysers. Moscow.

238

SUBJECT INDEX

Acetylcholine, content in neuroglia, 14 distribution in vacuoles, 8 effect on ISPO,52, 53 synthetic mechanism, 125 Acetylcholinesterase, hydrolysis of acetylcholine, 114, 115 Actomyosin, effect on protoplasmic movements, 125 Adenosine tr iphosphate, effect on CNS, 14 function in muscle contraction, 125 metabolism, regulated by feedback, 128, 129 Adenylic acid, occurrence in muscle contraction, 125 Adrenals, secretion, hypothalamus influence, 120 Adrenolytic effect, effect of chlorpromazine, 109 Afterpotential, recruitment of apical dendrites, 67 -69 Alcohol, effect on muscle rhythms, 132 Amygdaloid nucleus, relation to hypothalamus, 91 Amytal, effect on electrical parameter of cortex, 192 Anaemia, cerebral, result from increase intracranial pressure, 182 Anaesthesia, effect on ISPO,43 Antidiuretic hormone, content in cerebrospinal fluid, 120 production from "stress state", 93 Arcuate nucleus, effect of coagulation, 15 Area postrema, content, 5-hydroxytryptamine, 96 recording ISPO,73 Asphyxia, potential differences, brain, 27 relation to blood resistance, 34 Astrocytes, characteristic properties, 8

Atropine, effect on ISPO, 53, 113 effect on reflex excitability, 52 Autonomic ganglion, potassium loss, 158 Auto-oscillatory process, origin in, 73 Axon, course in CNS, 4, 5 giant, membrane, 139 slow rhythmic potential oscillation, 131 Barbiturate, effect on ISPO,53 mechanism of action, 192 Blood-brain b a r r i e r , interstitial fluid influences, 35 reticular formation mediator, 101 Blood pressure, effect of epinephrine, 182 effect on potential differences, 27 relation t o intracranial pressure, 204 relation to potential waves, 45 Brain stem, stimulation reticular formation, 37 Burst, brain s t e m activity, 75 spontaneous activity, 76 Caffeine, effect of intravenous injection, 185 relation to cortical impedance, 186 Carotid sinus, indirect action of epinephrine, 111 Catatonia, EEG recording, 196 increased resistance, cerebral cortex, 197 Catecholamine, location in brain areas, 111 Cerebral blood flow, effect on electrical parameters, 182 Cerebral cortex, ionic exchange, 198 Chloroform, application on ganglia, 199

SURJE CT INDEX

effect on ISPO,136 Chlorpr omazine, blocking action on reticular formation, 109, 122 Choliner gic neurons, cortex, recording ISPO, 83 Cholinergic synapses, presence reticular formation, 114 Cholinesterase, distribution in brain structure, 113 inactivation of acetylcholine, 13 inhibitory mechanism, 84, 89, 117, 120, 122 part of b a r r i e r , 14 Cocaine, administration to olfactory recept o r , 199 excitatory effect, 201 Convection, electrical activity, 72 Convulsive activity , changes in threshold, 32 ISPO,periodic shift, 61 Convulsive agents, cholinesterase inhibition, 187 Corpus callosum, distribution ISPO,96 D.C. potential, evaluation in cerebral cortex, 28 recording conditions, 26 Deafferentiation, cortex strips, electrical activity, 75 subcortical structures, 80 Dendrites, apical, characteristic features, 4, 5 dipole structure, 30, 33 effect on cortical activity, 17 resistance mechanism, 184 basal, content gemmules, 7 receptor afferent impulses, 17 excitatory effect on neuron, 36 hyperpolarization, 66 radiosensitivity , 207 synaptic structure, 137 Depolar iz a t ion, apical dendrites, synaptic stimulation, 16 excitatory stimuli, 12, 13 GABA inhibition, 15 influence from vacuolar, 14 relation to action potential, 11 De sympathiz ation, action similar to GABA, 72

239

cortex, impedance measurement, 50 Diencephalon, recording ISPO, 97, 98 DNA, content in nucleus, 10 Electroconductivity, effect on cortical dipole, 28 Electrocorticogram, indication of spike potential, 21, 22 potential oscillations, 60 Electrode, gain control, 24 recording ISPO,40 resistance microcapillary, 26 Electroencephalogram, indicator excitatory activity, 20 Electrostasis, effects of temperature, 167 Emotional reactions, activity of hypothalamus, 93 Emotions, hippocampal circuit, cortex relation, 141 Emotive background, conditioned reflexes, Pavlov, 94 Enzymatic processes, isolated cortex, potential fluctuations, 82 Enzymes, relation to rhythmic process, 128 substrate concentration, 126 Epinephrine, hypothalamic neuro-secretion, 43, 95 intravenous injection, reticular formation, 110 sympathomimetic agent, 130 Episodic bursts, brain stem, relation t o spontaneous activity, 75 Esterase, content of non-specific, 115 Ether, effect on cortical resistance, 192 Excitation, cortex strips, stimulation, 74 interaction of brain activity, 19 involvement of axodendritic synapses, 16 negative potential, cortical neurons, 33 summation of refractory period, 13 surface negative potential, 17 synchronization, 18

240

SUBJECT INDEX

Extinctions, ISPO,conditioned defense reflex, 142, 143 ISPO, cortex, 146 Factor I, cerebral cortex, inhibitory transmitter, 15 Factor P, gray substance, spinal cord, 14 Feedback, control mechanism, 128 giant axon, damping oscillations, 139 stability control, oscillation, 133 Fluid compartment, cerebral cortex, 9 Fornix cerebri, cerebral cortex, stimulation after hippocampal circuit, 92 Ganglion, autonomic, warm -blooded animals, 158 electrocorticogram, ISPO,50 resistance, ionic change, 200 secretory activity, 198 Giant axon, membrane potential stability, 131 resting potential, 132 Glia, involvement transport hormones, 136 rhythmic secretion, 73 similarity to secretory cells, 135 Glucose utilization brain slices, 183 Gray substance, brain stem, reticular formation, 98 substrate, electrical properties, 203 Hippocampal circuit, activity, 141 cortical coordination, 92 relation to thalamus, 92 Hippocampus, connection t o hypothalamus, 91 relation to wakeful state, 98 Histamine, ISPO,frequency of hypothalamus, 121 Hormones, antidiuretic, 93, 120 irradiation, 208

local chemical factor, 125 stimulation of hypothalamic area, 59 5 -Hydr oxy tr ypt amine, content in hypothalamus, 95 Hyperaemia, cerebral, relation t o intracranial pressure, 181 Hyperpolarization, effect of apical dendrites, 16 unmasking, amino-acid, 12 Hypophysis, connection to infundibulum, 151 Hypothalamus, effect on polarization, 32 excitation, 106 X-irradiation, 208 neurohumoral relationship, 91 regulatory system, 59 synchronous waves, after stimulation, 64 Impedance, dendritic area, cerebral cortex, 66 method of investigation, cerebral cortex, 156 radiation effect, 212 significance in synchronization, 18 Inductance, membrane component, 162 Infrastability, slow control system, 154 Inhibition, content of cholinesterase, 84, 89, 117, 120, 122 spontaneous activity, 17 synaptic activity, 11 Insulin, injection, effect on ISPO,,51 Internal capsule, ISPO distribution, 96 Inter neur on, contribution to dendritic activity, 69 Intoxication, infraslow oscillations, 57, 58 Intr a c r anial pressure, changes, relation to hemodynamics of brain, 204 electrical waves, cranial cavity, 44 Ionic changes, cerebral cortex, 198 effect on carbohydrate metabolism, 204 Ionic composition, metabolic mechanisms, 198 Ionic equilibrium, afferent impulses, resting state, 201

SUBJECT INDEX

,Ionizing radiation, effects on ISPO, 207 Ions, antagonism, 159 distribution, 156 exchange, 158 immobility, 157 membrane permeability, increase, 11 migration, 198 Isotope, surface penetration, giant axon, 156 Lactic acid, brain slices, stimulation, 183 effect from hypoxia, 180 Lamina terminalis, relation to hypothalamus, 152 Latency period, analyzer of excitability, 65 slow control system, 154 Luminal, effect on ISPO, 54 Mammillary bodies, connection with central gray substance, 92 Membr me, depolarization, 134 hyperpolar ization, 135 resistance measurement, 163 Metabolic tone, cerebral cortex, changes in levels of impedance, 198 changes in electrolytes, 205 Microsomes, acetylcholinesterase content, 114 enzyme content, 128 Mitochondria, cholinesterase content, 114 electron microscope; synapse location, 8 enzyme content, 128 Narcosis, blood-brain, 34 ECoG, magnesium, 61 ISPO changes, 54 Neurohormonal complex, organized behavioral act, 146 Neur ohor mones, ganglion, neurosecretory processes, 139 ISPO, central gray substance, 121

241

Neur onogr aphy , strychnine, methodology, 103 Neurosecretion, accumulation in hypothalamus, chemical characteristics, 123 nerve structures, 9 Norepinephrine, content in hypothalamus, 14, 95 Oligodendroglia, structure of cells, 8 Oxytocin, vasoconstr icting hormone, cerebrospinal fluid, 120 Permeability, excitatory postsynaptic potential, 157 Phosphacol, inhibition of cholinesterase, 75 intensification of ISPO, 117 parameter changes cortex, 84 relation to resistance and capacitance, 187 Phosphocreatine, content in brain during hypoxia, 180 Phosphorus, electrical stimulation of brain slices, 183 Physostigmine, cholinesterase inhibitor, 125 effect on giant axon, 133 Plasmodium, rhythmic changes in potential, 124 Polymer circuit, relation to freedom of rotation, 167 Potentials (see also Afterpotential; D.C. potential; Spike potential), conduction in axon, 13 involvement of dendrites, 10, 12, 16 68 membrane, 10 postsynaptic, 205 recruiting, 71, 72 resting, 20 Preoptic nucleus, effect of stimulation, 81 irradiation, 21 2 Procaine, inhibitor of neuron activity, 71 Protoplasm, potential oscillations, 124 rhythmic changes, 125 Pyramidal cells, excitability, stimulation cortical surface, 184

242

SUBJECT INDEX

Radiosensitivity, dendrites, 207 muscles, 207 Reflex, conditioned, 35, 37, 141, 154 defense, 142, 143 Refractory period, axons, electrical stimulation, 13 Resistance , asphyxiation, 179 barbiturates, 192 blood pressure, 181 caffeine, 186 convulsions, 187 effect of dissolved salt, 160 epinephrine, 179 membranes, 163 motor excitation, 195 stimulating drugs, 185 urethane, influence, 191 Respiration, wave synchrony, 44 Response, dendrites, 37 stimulation of thalamus, 31 Reticular formation, aperiodic shift of potential, 22 effect on cortex ISPO, 48 relation to slow regulatory systems, 59 repeated stimulation, cortical activity, 63 stimulation, relation to quasi-steady potential, 37 RNA, content in cytoplasm, 10 Satell it e cells, nutrients nerve metabolism, 9 termination of axons, 4 Secretion, presence in glia, 73 structure of granules, 123, 135 Seizure, motor excitation sound, 194 relation to brain resistance, 196 Semipermeability, polarization effect, 162 Serotonin, comparison topical application, 7 6 distribution in brain, 14 hypothalamus stimulation, 76, 88 Sleep, EEG, relation to atropine activity of hypothalamus, 93 fluctuation electrical parameters, 170

resistance, 190 Spike potential (see also Potentid), heterogenous membrane, 12 local excitation, cerebral cortex, 20 origin in initial segment, 11 postsynaptic potential, 21 spreading excitation, axon, 38 Spreading depression, action of potassium, 159 intensity of stimulation, 184 Stellate cells, afferent fibers, sensory pathway, 7 termination of axons, 4 Stress, antidiuretic hormone, 93 cortical response, 90 defense reaction, response, 48 ISPO,factors of influence, 58 reaction of body to injury, 53 Strychnine, comparison of effects to GABA, 51 EEG changes, 64 spontaneous activity, 78 Sube sophageal ganglion, inhibitory action, 203 motor function, 199 Substance P, content hypothalamus, 95 ISPO, hypothalamus, 121 Suppression, dendrite potential, 72 Supraoptic nucleus, stimulation, effect on ISPO, 87 Sympathetic ganglion, ISPO, effect of removal, 49 Sympathetic nervous system, tonic activity, nervous system, 49 trophic influence, muscles, 130 Synapses, action strychnine, 103 b a r r i e r to acetylcholine diffusion, 14 blockade of depolarization, 15 chemical transmitter, 13 current formation after depolarization, 11 distribution on dendrites, 137 inhibitory, 12 pericorpuscular, 7 spontaneous subliminal small potentials, 21 Synchronization, dendrite membrane, 17 EEG, 60 potential oscillation, cortical s t r u c tures, 18

SUBJECT INDEX

Thalamus, afferent pathways, 7 discharge, connection to hypothalamus, 103 effect on recruiting potentials, 72 radiation, 208 rhythmic stimulation, 18 stimulation, effect on cortex, 31 Tubocurarine, effect on dendritic potential, 69 Ultrastable system, control system, 153 Urethane, depressive effect on ISPO, 54 short-term effect on impedance, 191 Vasopr es sine,

243

glial secretion, 95 pituitary secretion, 120 Ventricle, artificial polarization, 32 -33 Veratrine, application to mollusk nerve, 131 inhibitory effects on neurons, 71 Wakefulness, activity of hypothalamus, 93 fluctuation electrical parameters, 179

X - ir r adiation, hypothalamus, 208

This Page Intentionally Left Blank