Search

A new concept of brain functions is proposed. According to it, the brain is a hierarchy of neural optimal control systems where each level possesses a dynamic model of its controlled object; this can be a lower control level in the CNS or an executive organ. The new concept of brain functions is based on the solution of the problem of central pattern generators and on the neural network computational principle. Contemporary data on the cerebellum, cortico-basal ganglia-thalamo-cortical loops (the structural/functional basis of the highest brain levels), Parkinson’s disease, and methods of its treatment (including deep brain stimulation) are analyzed from the point of view of the new concept of brain functions. These examples demonstrate the universal applicability of the new concept in the investigation of the physiology and pathophysiology of neural networks and the respective clinical effects.

In experiments on immobilized decerebrate cats, data about reorganization of efferent activity parameters of the forelimb and hindlimb locomotor generators evoked by electrical stimulation of descending systems were obtained. The generators controlling both forelimb and hindlimb locomotor movements were found to be characterized by the existence of stable states at which total influence of different descending systems on these generators was extremely limited. These data enable us to conclude that the sense of activity reorganization in locomotor generators of both forelimb and hindlimb under the influence of descending system signals is in bringing the motor program to a dynamic relation with supraspinal inflow, where a sufficient degree of limitation and balancing of the influences of corresponding descending systems on the interneuronal nets, determining time and phase characteristics of these generators, is ensured. Possible mechanisms of realization of this interaction between descending signals and locomotor activity generators are discussed.

Whole-cell glycine-activated currents were investigated in chick spinal cord neurons cultivated for up to three weeks. Based on the morphological and electrophysiological characteristics of neurons, two different types of nerve cells were distinguished during the first few days in culture. The first type consisted of “mature” nerve cells which appear to be motoneurons. They died by five to seven days in vitro . Immature neurons or neuroblasts constituted another type of nerve cell. They developed in culture and became differentiated neurons. Glycine-activated currents were elicited in both types of neurons during different periods in vitro . Sensitivity to glycine of “mature” neurons decreased from two to five days in vitro : ed 50 for agonist action increased from 0.4 to 1.3 mM. The sensitivity of neuroblasts to this transmitter increased during differentiation: ed 50 decreased from 1.4 to 0.12 mM on three to 14 days in vitro , respectively. Changes in glycine-activated conductance of these developing neurons were investigated later on. The conductance in differentiated neurons was markedly sensitive to membrane potential, while neuroblasts did not show such dependence. Voltage sensitivity was due to voltage-dependent kinetics of the ion channel. Desensitization kinetics of the glycine-activated currents were double-exponential. The time constant for the slow desensitizing component was dependent on glycine concentration, which was not the case for the fast component. The increase in glycine sensitivity of the neuroblasts was accompanied by deceleration of desensitization kinetics of the agonist-activated currents. A remarkable feature of the currents elicited in neuroblasts was their extremely long time course after rapid agonist removal from the cells. The properties of these long-term currents suggest that a large fraction of the receptors are desensitized, even during quite short applications of the transmitter. The presence of glycine in the culture medium did not affect the increase of neuronal sensitivity to the agonist. The block of spontaneous bioelectric activity by adding tetrodotoxin to the culture medium abolished developmental changes in glycine-activated conductance. Possible mechanisms for the changes in transmitter sensitivity of the neurons are considered.

Crystal structures of amino-acid receptor agonists glycine, taurine, GABA, and β-alanine were compared. Degree and accuracy of coincidences of distance between nitrogen and oxygen atoms in the crystal structures were used as criteria for similarity. It is concluded that crystal structure of agonists and agonist action on amino acid receptors are correlated. A theory is advanced that, when an agonist has a cooperative effect on a receptor, relative spacing of molecules on the receptor surface is similar to the spacing of molecules in the agonist's crystal structure.

The influences of signals in descending systems on the parameters of scratching generator activity were studied on decerebrate immobilized cats. It was shown that phasic electric stimulation of descending systems evoked certain phase-dependent reorganization of the parameters of scratching generator efferent activity. Maximum increase in scratching cycle duration during electric stimulation of Deiters' nucleus, red nucleus and pyramidal tract is observed during stimulation in the first half of aiming phase. Stimulation in the second half of aiming phase and at the beginning of scratching jerk phase virtually does not change the scratching cycle duration. Maximum increase in scratching cycle duration during electric stimulation of the nucleus reticularis gigantocellularis is observed in the second half of aiming phase. Electric activation of descending pathways during aiming phase increases its intensity and decreases the intensity of scratching jerk phase. Activation of descending pathways during scratching jerk phase increases its intensity and virtually does not change the aiming phase intensity. Influences of electric activation of descending systems on scratching generator work reveal dependence on limb position. They are increased when the limb is deflected to the rear and are decreased during over-aimed position. Decerebellation leads to a decrease of scratching generator activity parameters rearrangement under influence of electric stimulation of the red nucleus and nucleus reticularis gigantocellularis, and to its increase during Deiters' nucleus stimulation. On the basis of these results the principles of supraspinal correction of scratching generator work are discussed.

The effects of N-methyl-D-aspartate (NMDA) and the NMDA receptor antagonist, 2-amino-5-phosphonovaleric acid (2-APV), on spontaneous activity of dorsal and ventral roots (DR and VR, respectively) generated by isolated spinal cord from 16–20-day-old chick embryo were studied. This activity was synchronous oscillations of electrotonic potentials in DR and VR. There was no impulse activity in the VR. When NMDA was applied at 2–25 µM, the amplitudes of the oscillations increased, the impulse activities in VR and DR developed, and the tonic component of electrotonic potentials appeared. At 20 µM, 2-APV decreased both, the spontaneous and NMDA induced activity. After sectioning of the spinal cord, the neuronal network of the isolated dorsal arm conserved the capacity to generate spontaneous activity in the DR which increased after NMDA application. There was no rhythm in the ventral part of the spinal cord. The localization of the NMDA-sensitive neuronal network, generator of the rhythmic (motor) activity, in the spinal cord is discussed.

Effects of noradrenaline precursors on glycine and N-methyl-D-aspartate (NMDA) receptors in spinal cord neurons recently isolated from chick embryo were investigated using whole cell patch-clamp and concentration clamp techniques. Both L-alanine and L-DOPA were found to be glycine agonists capable of potentiating NMDA response, while L-tyrosine does not activate glycine but can potentiate NMDA response. Lastly, L-phenylalanine and dopamine do not interact with either glycine or NMDA receptors.

Effects of signals proceeding along descending pathways on parameters of scratch generator activity were investigated in decerebrate immobilized cats. Certain phase-linked alterations in these parameters were shown to occur under the effects of electrical activation of the main descending systems. The biggest increase in scratch cycle duration under electrical stimulation of Deiter's nucleus, the red nucleus, and pyramidal tract is produced when stimuli are applied during the first half of the aiming stage. Stimulation during the second half of the aiming phase and at the start of the scratching movement hardly affect the scratching cycle. The main increase in length of scratch cycle during electrical stimulation of the reticular gigantocellularis nucleus is noted when stimuli are presented during the second half of the aiming stage. Electrical activation of descending pathways during the latter induces a rise in intensity in this phase and reduced intensity of the actual scratching stage. Activation of the pathways during this scratching motion causes heightened intensity of the motion while hardly affecting intensity of the aiming phase. The principles of suprasegmental rectification of scratch generator operation are discussed.

The crystal structures of glycine, taurine, GABA, beta-alanine were compared. The quantity and the accuracy of distances coincidence between nitrogen and oxygen atoms were used as a criterion of similarity of the crystalline structures. The conclusion is made about a correlation between crystalline structure of agonists and their effect on amino acid receptors. It is assumed that in case of a cooperative effect of agonist on the receptor a mutual arrangement of molecules on the receptor surface is similar to their arrangement in the agonist crystal.

Pheochromocytoma PC12 cells incubated with and without nerve growth factor were investigated using the patch-clamp technique in the whole-cell recording mode, and the concentration clamp method in the rat. On the fourth day of incubation with nerve growth factor, sodium potential-activated ionic currents appeared in the membranes of the most morphologically differentiated cells. At the same period a three-fold increase of acetylcholine-activated current density, compared with the cells incubated without nerve growth factor, was observed. Thus, the qualitative and quantitative changes in membrane properties can be a result of metabolic reorganization in PC12 cells induced by nerve growth factor and accompanied by morphological differentiation according to neuronal phenotype.

The action of glycine and GABA on isolated lamprey spinal cord neurons was investigated by means of intracellular perfusion and concentration clamp techniques. These amino acids activated desensitizing chloride ionic conductances. The concentrations of agonists evoking half-maximum effects (ED50) were equal to 16 microM and 1.5 mM for glycine- and GABA-activated currents, respectively. Increase in the transmitter concentration led to a decrease in the time constant of desensitization. Current-voltage relationships of glycine- and GABA-activated currents were strongly dose-dependent. At low agonist concentrations the time constant of activation decreased with membrane hyperpolarization. Glycine- and GABA-activated currents exhibited complete cross-desensitization. The specific glycine antagonist, strychnine, suppressed both glycine- and GABA-activated currents to the same degree. Selective antagonists of GABA receptors, bicuculline and picrotoxin, produced equal blocking effects on glycine- as well as GABA-evoked responses. In the cells studied, taurine activated desensitizing ionic conductance. Responses evoked by taurine and glycine applications demonstrated complete cross-desensitization. Taurine-activated currents were sensitive to strychnine, bicuculline and picrotoxin. These results suggest the existence of one receptor-channel complex for the main inhibitory transmitters in lamprey spinal cord neurons. 3,4-Dioxy-L-beta-phenylalanine evoked desensitizing strychnine-sensitive ionic responses which exhibited cross-desensitization with glycine-activated currents.

Systematic quantitative analysis of changes in the spinal scratching generator motor activity evoked by tonic and phasic peripheral afferent signals during "fictitious" scratching was carried out in the cat. Correlations between the kinematics of hindlimb scratching movement, sensory inflow, and primary afferent depolarization were investigated. Reliable correlations between the parameters of generator motor activity during fictitious scratching were revealed: they depended on tonic peripheral afferent inflow. The functional role of these dependencies consists of providing stability for aiming the hindlimb to the itch site. It was shown that scratching generator reaction to a phasic sensory signal depended significantly on afferent input, signal intensity, and its arrival phase in the cycle of motor activity. Phase correction of "scratching" rhythm was performed by inhibition of the current stage of "scratching" cycle, the inhibition magnitude depending on the intensity of a sensory signal run along high threshold afferent fibers. The moments in the scratching cycle, in which the afferent signal caused no rearrangement in scratching generator activity, were discovered for all investigated afferent inputs. These moments corresponded to the transitions from one scratching cycle phase to another. Integral afferent activity was distributed unevenly in the cycle during real scratching. The main part of it was observed just in that scratching cycle part which included the above mentioned no rearrangement phase points. The data obtained allowed us to conclude that the scratching generator should be considered as a working program for the motor optimal control system containing the intrinsic model of the controlled object dynamics (e.g. hindlimb scratching movement dynamics), which produces an inner analog of peripheral flow. This inner flow interacts with peripheral afferent inflow just as one of the latter components. Centrally originated modulation of primary afferent depolarization is a result of affecting the depolarization generating system by this inner "sensory" activity. It is the model, with the aid of which the generator can work after deafferentation. The functional organization of a central pattern generator is considered.

Bath-application of N-methyl-D-aspartate (NMDA) and antagonist of NMDA-receptors 2-amino-5-phosphonovaleric acid (2-APV) has been studied from the standpoint of its effect on the spontaneous activity in dorsal and ventral roots (DR and VR, respectively) generated by isolated spinal cord of 16-20-day-old chick embryo. Spontaneous activity consisted of synchronous oscillations of electrotonic potentials in DR and VR. Bath-application of NMDA (2-25 mumol/l) increased amplitude of the electrotonic potentials, induced spice discharges in DR and VR, a tonic component of electrotonic potentials. Bath-application of 2-APV (20 mumol/l) depressed spontaneous and NMDA-induced activity. The neuronal network of isolated dorsal horn (after splitting of the spinal cord) retained ability to generate spontaneous activity in DR which was intensified after bath-application of NMDA. No rhythmic activity appeared in the ventral part of the cord. The location of NMDA-sensitive neuronal network generating rhythmic (motor) activity in the spinal cord is discussed.

1. The membrane currents elicited by L-glutamate and glycine applications in morphologically different neurone types were investigated in isolated spinal cord cells from the lumbar enlargement of 6 to 11-day-old chick embryos. The whole-cell patch-clamp technique and concentration clamp methods have been used. Isolated spinal cord neurones of four stages were investigated: 6, 7.5, 9 and 11 days of incubation (29th, 32nd, 35th and 37th stages of development, respectively). 2. The L-glutamate-activated conductance consisted of desensitizing and non-desensitizing components. The Hill coefficient for the first component was 1, and for the second was 2. The number of cells responding to L-glutamate application with only a desensitizing component decreased from 53.4% on the 6th day of incubation to 6.7% on the 11th day, whereas the number of cells responding with bicomponent responses increased during the same period from 13.3 to 87%. 3. From the 6th and 11th day of chick embryo development the characteristics of the desensitizing component of L-glutamate-activated conductance remained constant (half-maximal dose, (ED50 = 2.6 +/- 0.3 mM) whereas the ED50 for the non-desensitizing component decreased 10 times. 4. It was found that the density of the desensitizing L-glutamate-activated ionic current increased during morphological and age-dependent differentiation of spinal cord cells. 5. Of the investigated cells 88.7% were sensitive to glycine application. The smallest percentage of neurones responding to glycine application was observed at 7.5 days of incubation. Glycine-activated conductance did not change at the investigated stages (ED50, 71 +/- 2 microM; Hill coefficient, 2). 6. A significant decrease in the glycine-activated current density was observed on the 9th day of incubation in multipolar neurones with three neurites, against a tendency to an increase of glycine-activated ionic currents during morphological and age-dependent differentiation of cells. 7. Of the investigated cells 78% were sensitive to both glycine and L-glutamate application. A negative correlation between glycine- and L-glutamate-activated current densities (correlation coefficient, -0.71) was revealed by means of statistical analysis. 8. We conclude that these changes in chemosensitivity of the spinal cord neurones may underlie the increase in locomotor activity of the chick embryo observed on the 9th day of incubation.

Effect of noradrenaline precursors (L-alanine, L-phenylalanine, L-tyrosine, L-DOPA and dopamine) on glycine and NMDA receptors of freshly isolated spinal neurones of chick embryo were investigated by means of the whole cell patch-clamp and concentration clamp techniques. L-alanine and L-DOPA were found to be glycine agonists which can potentiate NMDA responses. L-tyrosine does not activate glycine receptor but potentiate NMDA one. L-phenylalanine and dopamine do not interact with both glycine and NMDA receptors.

Rearrangement of the scratching generator activity evoked by phasic electrical stimulation of different descending systems has been investigated on decerebrated immobilized cats. This rearrangement heavily depends on the stimulation phase. Maximal increase of the scratching cycle duration by electrical stimulation of Deiters' nucleus, red nucleus and pyramidal tract takes place in the first half of an aiming phase. The electrical stimulation of structures mentioned above does not practically change duration of the scratching cycle in the second half of the aiming phase and at the beginning of the scratching phase. A maximal increase of the scratching cycle duration by electrical stimulation of reticular gigantocellular nucleus occurs in the second half of the aiming phase. The electrical activation of descending pathways during the aiming phase increases intensity of this phase and decreases intensity of the scratching phase. The electrical activation of descending pathways during the scratching phase increases its intensity and does not change practically the intensity of the aiming phase. Possible principles of the suprasegmental correction of the scratching generator activity are discussed.

A new theory of the cerebellum work is presented. It explains all the experimental data now accumulated.

The relationship between parameters of electrical muscle activity, changes at hindlimb joint angles, intensity of integral afferent flow, and dorsal root potential during real-life locomotion was investigated in cats decerebrated at high level. Characteristics of rear limb movements before and after deafferentation were described. It was found that afferent activity during locomotion motion consists, of two components — a tonic and a periodic phasic stage. Three main waves may be distinguished in the latter, each of which gives rise to associated changes in the level of primary afferent terminal polarization. These changes in turn are summated with the effects produced by the central generator. Correlations, between the parameters of these processes were investigated and the mechanisms underlying afferent control of locomotion generator function discussed.

Sections of the medulla known as the locomotor strip were investigated during cat experiments using the horseradish peroxidase retrograde transport technique. The enzyme was administered under functional control (stimulating the injection site evoked locomotion). Results showed no evidence of a hypothetical column of cells either along the locomotor strip or medially to the strip. This would cast doubts on the existence of such a strip as a separate structural formation. It was shown by comparing findings from morphological research and experiments involving electrical stimulation that the bulbar locomotor strip may in fact consist of the spinal trigeminal tract and its nucleus. It is postulated that non-specific afferent activation of the brainstem reticular formation plays a crucial role in initiating locomotion.

Pyramidal unit activity in unanesthetized cats at rest and during voluntary movement was recorded by a microelectrode technique from the motor cortex for the forelimb. Some pyramidal neurons were not spontaneously active. The conduction velocity along the axon of these neurons was sometimes high (up to 71.5 m/sec), sometimes low (up to 11.2 m/sec). The remaining pyramidal neurons had spontaneous activity with a mean frequency of 1.29 to 43 spikes/sec. Analysis of interspike interval histograms of spontaneous activity and of autocorrelation histograms showed grouping of the spikes into volleys in most pyramidal neurons (irrespective of the conduction velocity). During voluntary movements the change in the activity of many pyramidal units correlated with changes in the EMG. The firing rate of the pyramidal neurons under these circumstances began to rise at least 50 msec before the increase in amplitude of the EMG and it remained high throughout the movement. The firing rate of most neurons during movement was 40–60/sec. The results are compared with those obtained by other workers who studied pyramidal unit activity of monkeys during voluntary movement.

Using unanesthetized and decorticated (or decerebrated at level A 13) cats, it was found that spinalization leads to depolarization of the central terminals of primary afferents and an increase in the N1 component of dorsal surface potential and dorsal root potential (DRP) produced by stimulating the low-threshold cutaneous and muscle afferents. Other effects include an increase in early polysynaptic responses and DRP produced by stimulation of high-threshold muscle afferents, a reduction in the intensity of interneuron activation in the nucleus interpositus mono- and polysynaptically connected with primary afferents, and a rise in the activity of n. interpositus interneurons di- and oligo-synaptically connected with afferent terminals. Changes in the opposite direction were produced by injecting DOPA into spinal animals. The connection between changes in the state of the segmental neuronal apparatus of the lumbosacral spinal cord and the level of spinal locomotor generator activity is discussed in the light of the findings obtained.

Reorganization of the parameters of efferent activity produced in the spinal generator by electrical stimulation of the ipsilateral hindlimb muscle nerves during different limb positions were investigated in decerebrate immobilized cats. A direct relationship was found between this reordering and the stage at which stimuli were applied. The rearranged duration of the scratch cycle showed a tendency to bring motor activity into phase with stimuli so that the stimulus falls due at the onset of the motor activity phase. This phasically collated rearrangement was observed where a shift had occurred in the relationship between "aiming" and "scratch" motion tending towards intensified activity in the muscles innervated by the stimulated nerve. Rearrangement became more evident when the hindlimb deflected from the target position in accordance with the direction of muscle stretching. The physiological significance of the interposition of the "no rearrangement" phase is discussed. It is deduced that this absence of change in duration and intensity can only be produced simultaneously when a certain relationship is achieved between the phase of afferent signal reception in the scratch cycle and signal intensity.

The number of spikes which must be recorded in order to detect significant correlation between spike trains of two synaptically connected neurons was estimated by a mathematical model. Dependence of this number of spikes on importance of interneuronal connection (measured as the amplitude of the EPSP evoked by a single spike of the input neuron in the output cell) and on the intensity of total spontaneous excitatory influences on the output neuron and on its own parameters was studied. For cells which corresponded in the weight of connections between them, their intrinsic parameters, and characteristics of spontaneous activity to real spinal neurons, the necessary number of spikes was 107–108. An increase in amplitude of the single EPSP and also a decrease in the intensity of the input spontaneous spike train and parameters of after-hyperpolarization of the postsynaptic neuron led to a decrease in the number of spikes necessary for the detection of significant correlation. On the basis of the results of this and previous investigations the possible principles for construction of a spinal locomotor generator are discussed.

Unit activity of the lumbar interneurons was recorded in thalamic cats during fictitious locomotion. Neurons whose activity was modulated in the rhythm of fictitious locomotion were found in the lateral parts of the intermediate zone of gray matter and ventral horn. Of these neurons, 41.2% were activated mainly in the phase of "flexion," 48.5% in the phase of extension, and 10.3% in both phases. Neurons with tonically increasing or decreasing activity during rhythmic discharges and neurons whose activity was unchanged during fictitious locomotion also were observed. During later discharges all these neurons were similarly activated, although a depth of modulation of unit activity was lower than during fictitious locomotion. Afferent inputs to the recorded interneurons also were studied. The neuronal organization of the spinal locomotor generator is discussed on the basis of these results.

Changes in electrical polarization of primary afferent terminals in the lumbosacral portion of the spinal cord were investigated during fictitious locomotion in immobilized decorticated and spinal cats. Fictitious locomotion was accompanied by stable hyperpolarization of the afferent terminals, against the background of which they were periodically depolarized in rhythm with efferent activity. These tonic and phasic changes were observed in terminals of all groups of afferent fibers tested: cutaneous and muscular (Ia and Ib). Periodic in-phase depolarization was carried out in different ipsilateral segments of the lumbosacral enlargement. During ficitious galloping changes in depolarization of the primary efferents were in phase on different sides; during fictitious walking, these periodic changes were out of phase. On the basis of these results the physiological importance of changes in electrical polarization of primary afferent terminals of the spinal locomotor generator is discussed.

It was found during experiments on immobilized decerebrate (at intracollicular level) cats that tonic sub-threshold activation of the spinal generator of scratching action (following application of tubocurarine or bicuculline to segments C1-C2) was accompanied by depolarization of primary afferent terminals, a reduction in the N1 component of dorsal surface potential produced by stimulating the cutaneous afferents, and a reduction in the amplitude of dorsal root potentials and lead-phase polysynaptic response produced in motoneurons by stimulating the cutaneous and muscle afferents. A rise or a reduction in the activity of interneurons belonging to the interstitial nucleus connected respectively mono- and di-(oligo)synaptically with the afferents occurred in parallel with this. Spinalization produced the same changes in reverse in the animal. By administering DOPA to the spinal animal, a comparison could be made of changes occurring in the state of the segmental apparatus of the lumbar section of the spinal cord during tonic sub-threshold activation of spinal scratch generators and locomotor movements.

Activity of lumbar spinal neurons was recorded extracellularly during late long-lasting discharges in efferent nerves in immobilized thalamic cats. Of the total number of cells tested, 70% changed their activity during late discharges. The activity of 35% of neurons was increased during late discharges in nerves to flexors, but inhibited during discharges in nerves to extensors. Responses of 27% of neurons were of the opposite character. Other neurons were found whose activity was increased (5%) and reduced (3%), respectively, during later discharges in both flexor and extensor nerves. Most interneurons which changed their activity during late discharges were located in lateral parts of the intermediate zone of gray matter and the ventral horn at a depth of 2.8 mm. The character of the afferent input to a neuron was found to depend on the late efferent discharges and activity of the neurons correlated with them. Neurons whose activity was unchanged during late discharges (30%) were mainly located rather more dorsally, at a depth of about 2.0 mm. The possible mechanisms of the participation of these groups of interneurons in the generation of late discharges are discussed.

Studies on immobilized decerebrate (at intracollicular level) cats in which the scratch generator had been set up following bicuculline application to the upper cervical segments of the spinal cord, showed that the state of the segmental apparatus of the lumbosacral section of the spinal cord differs substantially from that seen in the spinal animal. Direct excitability of motoneurons of the "aiming" and "scratching" muscles rises, while recurrent and reciprocal Ia inhibition of motoneurons intensifies and the influence of Ib afferents on motoneurons declines. Afferents of the flexor reflex exert a primarily inhibitory influence on motoneurons of the "aiming" muscles. This influence becomes predominantly excitatory following spinalization, while the inhibitory effects of these afferents on motoneurons of the "scratch" muscles declines. The functional significance of the changes discovered in generation of scratch routine is discussed.

A statistical analysis of unit activity in spinal locomotor centers was undertaken on immobilized thalamic cats at rest and during generation of efferent discharges. Activation of the spinal locomotor generator was accompanied by shortening of interspike intervals in the spike sequences of neurons and a decrease in their fluctuations. Histograms of interspike intervals became more symmetrical under these circumstances and there was a considerable increase in the number of neurons whose activity showed regular fluctuations on autocorrelation histograms. Spike trains at rest were characterized by dependence of successive intervals, which increased during efferent discharge generation. The possible mechanisms of modification of the time structure of unit activity in spinal locomotor centers during their activation are discussed.

This paper summarizes information obtained in the experimental study of the dynamics of polarization of central primary afferent endings and modifications of segmental responses to afferent stimuli during fictitious locomotion and fictitious scratching in immobilized, decorticated, decerebrate, and spinal cats. Fictitious locomotion was accompanied by tonic hyperpolarization, fictitious scratching by tonic depolarization of central primary afferent endings. Against the background of these long-lasting changes in primary afferent depolarization, it exhibited periodic changes in the rhythm of efferent activity. Periodic changes of depolarization were virtually in phase in different ipsilateral segments of the lumbosacral enlargement. Data on groups of afferent fibers in whose central endings tonic and phasic changes of polarization took place. The appearance of fictitious locomotion was accompanied by a tonic increase, and of fictitious scratching by tonic inhibition of several evoked segmental responses. These tonic changes were a background against which segmental responses were modulated in step with the working rhythm of the locomotion and scratching generators. Many of the changes in evoked segmental responses were shown to be based on modulation of polarization of central endings of primary afferents by locomotion and scratching generators. It is concluded that active tonic and phase-dependent selection of incoming afferent information is effected through modulation of presynaptic inhibition of the generator. The role of this selection in peripheral collection of activity of locomotion and scratching generators is discussed.

Transient and steady-state components of L-glutamate-activated membrane currents were investigated using intracellular perfusion, voltage clamp, and concentration clamp techniques in spinal cord neurons of 6–11 day chick embryos. Hill's coefficient was found to equal 1 for transient and 2 for steady-state components. It was shown that the L-glutamate-activated receptors are present, which appear in the membrane of spinal neurons at the early stages of development.

Effects of stimulation of flexor reflex afferents were studied in decerebrate immobilized cats. Stimulation of ipsilateral afferents evoked late long-lasting discharges in the nerves to the flexors, whereas stimulation of contralateral afferents led to similar discharges in nerves to both extensors and flexors. Compared with spinal animals, early segmental reflexes in thalamic cats were tonically depressed. Similar tonic inhibition of segmental reflexes took place in spinal animals after injection of dopa. Segmental reflexes were clearly modulated during late or rhythmic discharges. The possible central mechanisms of these changes in the segmental reflexes are discussed on the basis of data in the literature.

Depolarization of primary afferent terminals in the lumbosacral portion of the spinal cord evoked by selective activation of propriospinal pathways was investigated in anesthetized cats. The strongest depolarization developed as a result of activation of short (two to five segments) propriospinal pathways in the lateral funiculus; stimulation of the long propriospinal pathways of this funiculus also induced depolarization, but of lower amplitude. Stimulation of propriospinal pathways of the ventral funiculi was ineffective. Significant primary afferent depolarization developed only following the use of a series of stimuli and strong stimulation of the propriospinal pathways. Excitation of these pathways caused depolarization of afferent terminals of both cutaneous and muscular nerves, including muscular sensory fibers of group Ia, although in the latter case its intensity was low. Neuronal mechanisms involved in the generation of this depolarization and its possible functional role are dicussed.

The influence of tonic afferent inflow as conditioned by ipsilateral hindlimb position on the efferent activity parameters of the spinal generator governing scratching motion was investigated in immobilized decerebrate cats. A significant correlation was observed between motor activity parameters and ensuing bouts of scratching in the absence of afferent flow (after deafferentation of the limbs). This correlation was less pronounced when afferentation remained intact and declined when the limb was shifted from the "aimed" to either the "overaimed" or "deflecting backwards" placing of the limb. The statistically significant correlations found between the parameters of different stages of motor activity and their dependence on hindlimb positions during actual scratching could be responsible for the stability of intended placing of the limbs during the performance of oscillatory movements. Hindlimb deafferentation would appear closest to "aimed" position judging by the parameters of efferent activity and the nature of correlations between them.

Segmental reflex responses of the lumbosacral region of the spinal cord to stimulation of peripheral afferents were studied in immobilized decerebrate cats before and after application of D-tubocurarine or bicuculline to the superior cervical segments, potentiating the scratch reflex, and also during fictitious scratching evoked by mechanical stimulation of the ear. Application of these substances led to inhibition of the N1-component of the dorsal cord potential, the dorsal root potential, and polysynaptic responses in efferent nerves. The appearance of fictitious scratching was accompanied by additional tonic inhibition of these responses, against the background of which modulation of the amplitudes of the responses was observed depending on the phase of fictitious scratching. Modulation of amplitudes of monosynaptic reflexes also developed during fictitious scratching. Against the background of these results the mechanisms and physiological role of reorganization of segmental responses during activation of the spinal scratching generator are discussed.

The pattern of change in the density of L-glutamate- and glycine-activated ionic currents through the spinal cord neuronal membrane of 6–11 day chick embryos was traced using an enzymatic technique for isolating cells during voltage-clamp experiments. A general trend towards increased current density was found during cell differentiation, followed by a substantial reduction in the density of glycine-activated current on the 9th day of incubation. A reciprocal relationship between densities of L-glutamate- and glycine-activated currents were characteristic of individual neurons. The functional significance of the patterns identified is discussed in connection with the evolution of spinal cord motor function during embryogenesis.

Research was performed on spinal cats injected with DOPA and decorticate (decerebrated at level A 13) and spinal cats. It was found that formation (activation) of the spinal locomotor generator is accompanied by heightened excitability in the extensor and the reverse trend in the flexor motoneurons, by an increase in the efficacy of recurrent and reciprocal Ia inhibition of α-motoneurons, and by a weakening of the influence of Ib afferents and extensor reflex afferents on these same motoneurons. The likely functional role of these changes in tuning of the spinal segmental apparatus in the generation of locomotor rhythm is discussed.

Activity of lumbosacral spinal interneurons was studied during fictitious scratching in decerebrate, immobilized cats. Neurons whose activity changed during fictitious scratching were located in the substantia intermedia lateralis and ventral horn. Among these neurons cells were distinguished whose activity was modulated in rhythm with motor discharges to different muscles (61.6%) and cells which were activated tonically (21.4%) or inhibited tonically (17%). By correlation of activity with discharges to corresponding muscles the rhythmically activated neurons were divided into "aiming" (36.6%) and "scratching" (25%). Neurons whose activity was unchanged during fictitious scratching also were observed. These cells were located mainly in the more dorsal regions of gray matter. Neurons to which wide convergence of excitatory influences from high-threshold cutaneous and muscular afferents was observed were mainly placed in the "aiming" group. "Scratching" neurons, compared with "aiming," more often received inputs only from low-threshold cutaneous or high-threshold muscular afferents. Group Ia interneurons were activated in phase with the corresponding motoneurons. Passive displacement of the limb in a forward direction predominantly inhibited spike activity of the "aiming" and potentiated activity of the "scratching" neurons. The neuronal organization of the spinal scratch generator is discussed on the basis of the results.

Segmental responses of the lumbosacral region of the spinal cord to peripheral afferent stimulation were studied in decorticated, immobilized cats before and during fictitious locomotion. The appearance of fictitious locomotion was accompanied by a tonic increase in the N1-component of the dorsal cord potential and dorsal root potential. Against the background of this tonic increase, modulation of these responses depending on the phase of fictitious locomotion was observed. When the N1-component and dorsal root potential were evoked at the end of the "extension" phase and at the beginning of the "flexion" phase their amplitude was greater, but when they were evoked at the end of the "flexion" phase and the beginning of the "extension" phase it was smaller. Polysynaptic and monosynaptic reflex response of motoneurons exhibited the same phase dependence during fictitious locomotion. The mechanisms and physiological importance of this retuning of segmental responses are discussed.

The kinetics of primary afferent depolarization (PAD) in the lumbar spinal cord was studied in immobilized decerebrate and decapitated cats during fictitious scratching. Fictitious scratching was accompanied by tonic and periodic PAD. Periodic PAD was cophased in different segments of the lumbosacral enlargement. Tonic depolarization was observed in terminals of all groups of afferent fibers studied (low-threshold cutaneous and of groups Ia and Ib). Periodic changes were observed only in endings of low-threshold cutaneous fibers and group Ib fibers. The physiological role of modulation of PAD by the spinal scratch generator is discussed.

The characteristics of conduction of the excitation wave along propriospinal fibers of the dorsolateral tract of the spinal cord were studied in cats anesthetized with pentobarbital. At a preliminary operation, 10–18 days beforehand, lateral hemisection of the spinal cord was performed, cranially in the lumbar division and caudally and cranially in the cervical division to the segments to be studied, leading to degeneration of the long descending and ascending fibers. During stimulation, the dorsolateral tract developed a composite response consisting of a positive-negative wave recorded up to 60–65 mm (4 or 5 segments) from the point of stimulation. The mean conduction velocity of this wave in the lumbar division was 37.9 m/sec compared with 44.5 m/sec in the cervical division. From its properties as a whole this wave can be regarded as the result of excitation of relatively fast-conducting propriospinal fibers of the dorsolateral tract. If the strength of stimulation was increased, late components began to appear in the response. These were evidently connected with excitation of thinner propriospinal fibers and synaptic activation of other other groups of spinal neurons.

Synaptic responses of 121 identified cervical motoneurons to stimulation of the pyramidal tract and red nucleus were investigated by intracellular recording in cats. Responses of EPSP or EPSP-IPSP type were predominant in motoneurons of distal groups of muscles and proximal flexors, while responses of IPSP type were predominant in motoneurons of the proximal extensors. The minimal effective number of stimuli for most motoneurons was 2 or 3. The mean latent period, counted from the first stimulus in the series, was 7.86 msec for EPSPs for stimulation of the pyramidal tract and 7.91 msec for stimulation of the red nucleus, while the corresponding periods for IPSPs were 8.68 and 8.75 msec. The segmental delay of 1.3–2 msec for EPSPs and IPSPs generated in certain motoneurons in response to stimulation of both structures indicates that the shortest pathway for transmission of activity from the fibers of these tracts to the motoneurons may be disynaptic. At the same time, the possible presence of an additional neuron for most inhibitory pathways cannot be ruled out. Analysis of the results also suggests the presence of a common interneuronal apparatus for both systems.

Responses of motoneurons and interneurons of the cervical enlargement of the cat spinal cord were studied by a microelectrode technique during selective stimulation of propriospinal fibers of the dorsolateral tract of the lateral white column. The long descending and ascending pathways were blocked by preliminary (10–16 days earlier) hemisection of the spinal cord cranially and caudally to the segments studied. Stimulation of the dorsolateral tract at a distance of 15–25 mm from the site of recording evoked complex postsynaptic potentials consisting of several successive waves in the motoneurons. The character of the PSPs was not clearly linked with the function of the motoneurons. By their latent periods the components of the PSPs could be placed in three groups. The "primary" components were reproduced in response to stimulation at 50–100/sec whereas the "secondary" and "tertiary" components were weakened or blocked. It is postulated that the "primary" components are evoked through monosynaptic connections between propriospinal fibers of the dorsolateral tract and motoneurons of the forelimb muscles, while the late components are evoked through polysynaptic pathways, including segmental interneurons. Many of these interneurons, located in the ventral horn and intermediate zone, were strongly excited during stimulation of the dorsolateral tract.

The distribution of focal potentials over the cross section of the 7th cervical segment of the spinal cord was studied during stimulation of the pyramids, the red nucleus, and a peripheral nerve (ulnar) in adult cats anesthetized with chloralose and Nembutal. The earliest focal potentials in the fasciculus dorsolateralis were recorded 1.4–1.5 msec after stimulation of the pyramids and 0.8–0.9 msec after stimulation of the red nucleus. These times correspond to maximal condution velocities of 56–68 and 105–124 m/sec respectively. The earliest post-synaptic activity in response to pyramidal stimulation was found in the lateral areas of laminae V and VI, and in response to stimulation of the red nucleus in laminae VI and VII in Rexed's classification. The pyramidal wave also evoked considerable postsynaptic activity in medial areas of the dorsal horn. In response to stimulation of peripheral afferents activity was evoked in neurons in the central and medial parts of laminae V and VI. It is postulated on the basis of these results that corticospinal and rubrospinal fibers may be connected monosynaptically with specialized interneurons, free from peripheral influences, in the lateral areas of laminae V and VII respectively; in the lateral part of lamina VI convergence of both types of influences on the same cells is possible. Interaction between descending and afferent influences possibly takes place on more medially located neurons.