Pathological Correlates of Cognitive Decline in Parkinson's Disease: From Molecules to Neural Networks

Parkinson's disease (PD) is a progressive neurodegenerative disorder caused by the death of dopaminergic neurons in the substantia nigra and appearance of protein aggregates (Lewy bodies) consisting predominantly of α-synuclein in neurons. PD is currently recognized as a multisystem disorder characterized by severe motor impairments and various non-motor symptoms. Cognitive decline is one of the most common and worrisome non-motor symptoms. Moderate cognitive impairments (CI) are diagnosed already at the early stages of PD, usually transform into dementia. The main types of CI in PD include executive dysfunction, attention and memory decline, visuospatial impairments, and verbal deficits. According to the published data, the following mechanisms play an essential role demonstrates a crucial importance in the decline of the motor and cognitive functions in PD: (1) changes in the conformational structure of transsynaptic proteins and protein aggregation in presynapses; (2) synaptic transmission impairment; (3) neuroinflammation (pathological activation of the neuroglia); (4) mitochondrial dysfunction and oxidative stress; (5) metabolic disorders (hypometabolism of glucose, dysfunction of glycolipid metabolism; and (6) functional rearrangement of neuronal networks. These changes can lead to the death of dopaminergic cells in the substantia nigra and affect the functioning of other neurotransmitter systems, thus disturbing neuronal networks involved in the transmission of information related to the regulation of motor activity and cognitive functions. Identification of factors causing detrimental changes in PD and methods for their elimination will help in the development of new approaches to the therapy of PD. The goal of this review was to analyze pathological processes that take place in the brain and underlie the onset of cognitive disorders in PD, as well as to describe the impairments of cognitive functions in this disease.

Neuronal aspects of septo-hippocampal relations

In unanaesthetized, conscious rabbits, in unstressful conditions, the neurons of the hippocampus and septum were investigated extracellularly during the presentation of a series of varied sensory stimuli. In the normal hippocampus these stimuli evoke habituating reactions of tonic (more usually, inhibitory) type in field CA3, with the addition of 'specific' patterned, and phasic reactions in field CA1. After complete septo-hippocampal disconnection the proportion of tonic (especially, of inhibitory) reactions in the hippocampus decreases. Theta bursts in the neuronal activity are absent; reactions to repeated sensory stimuli do not habituate. After lesion of the cortical perforant path to the hippocampus the majority of reactions in both fields are of tonic type. The proportion of neurons with regular theta bursts increases. Habituation is completely absent. A high correlation appears between the sensory reactions and the effects of midbrain reticular formation stimulation in the same neurons. The combination of both lesions does not significantly change the spontaneous activity of hippocampal neurons (except for the absence of the theta bursts). An increase in the level of activity of hippocampal neurons (by physostigmine), or rhythmic stimulation of the remaining synaptic systems, does not restore their rhythmic theta activity. In the septum deprived of hippocampal input the normal level of reactivity to sensory stimuli and the normal types of reaction are preserved. The proportion of neurons with theta bursts increases. The typical linear and rapid habituation of reactions disappears and is replaced by an unlimited increment in effects during repeated presentations of sensory stimuli. Discussion concerns the synchronizing and inhibitory influences of the septum on the hippocampus, and the role of the hippocampus in the organization of decremental processes (habituation) in the septum and brainstem structures.

[Theta rhythmicity in the medial septum: entraining by the GABA-ergic neurons]

The medial septal/diagonal band complex (MS/DB) is believed to play an important role in the generation and maintenance of the hippocampal theta rhythm, which has been implicated in the mnemonic and information-processing capacity of the brain. Although the physiological and morphological diversity of the septal neurons indicates their different functions, it is not known which cell type within the population contributes most critically to the theta rhythm. Here we review the chemical identity of different cell groups within the MS/DB complex, the anatomical connectivity between them, the electrophysiological properties of immunochemically-defined cell types, and their contribution to theta rhythmicity in the medial septum and the hippocampal theta rhythm. In order to better understand the mechanisms involved in rhythmic burst firing of the MS/DB neurons, a number of relevant theoretical models related to the generation/synchronization in neural networks are discussed.

Modulation of theta rhythmicity in the medial septal neurons and the hippocampal electroencephalogram in the awake rabbit via actions at noradrenergic alpha2-receptors

The modulation of the firing discharge of medial septal neurons and of the hippocampal electroencephalogram (EEG) mediated by actions on alpha2-adrenoreceptors (ARs) was investigated in awake rabbits. Bilateral i.c.v. infusion of a relatively low dose (0.5 microg) of the alpha2-AR agonist clonidine produced a reduction in the theta rhythmicity of both medial septal neurons and the hippocampal EEG. In contrast, a high dose of clonidine (5 microg) increased the percentage and degree of rhythmicity of theta bursting medial septal neurons as well as the theta power of the hippocampal EEG. On the other hand, administration of alpha2-AR antagonist idazoxan produced the opposite dose-dependent effect. While a low dose of the antagonist (20 microg) produced an increase in both the theta rhythmicity of medial septal neurons and the theta power of the hippocampal EEG, a high dose (100 microg) caused a reduction of theta rhythmicity in both the medial septum and hippocampus. These results suggest that low doses of alpha2-ARs agents may act at autoreceptors regulating the synaptic release of noradrenaline, while high doses of alpha2-ARs drugs may have a predominant postsynaptic action. Similar results were observed after local injection of the alpha2-AR drugs into the medial septum suggesting that the effects induced by the i.c.v. infusion were primarily mediated at the medial septal level. We suggest that noradrenergic transmission via the postsynaptic alpha2-ARs produces fast and strong activation of the septohippocampal system in situations that require urgent selective attention to functionally significant information (alert, aware), whereas the action via the presynaptic alpha2-ARs allows a quick return of the activity to the initial level.

Muscarinic blockade slows and degrades the location-specific firing of hippocampal pyramidal cells

The firing of rat hippocampal pyramidal cells is determined both by the animal's location and by the state of the hippocampal EEG. Because cholinergic transmission plays a role in EEG activity, we expected that its modification would alter place cell activity. We therefore investigated the effects on place cell activity of blocking muscarinic transmission with intracerebroventricular injections of scopolamine. Scopolamine reduced both the rate of place cell discharge inside firing fields and the spatial coherence of the fields; discharge outside of the fields also showed small increases. After injections, fields were shifted farther from their previous location than for saline controls, indicating reduced reproducibility after muscarinic blockade. Scopolamine increased the time rats were stationary, but changes in place cell activity persisted even after analysis was restricted to periods of walking, suggesting that the behavioral changes cannot account for the cell discharge changes. The scopolamine effects were dose dependent to an extent that varied between different measures. The firing rates of interneurons showed only a minor trend to decrease after scopolamine. Nevertheless, the spatial coherence of interneuron firing patterns was reduced, consistent with the recent demonstration that their positional firing is mediated by the location-specific firing of pyramids (Marshall et al., 2002). These results demonstrate that acetylcholine enhances positional firing patterns in the hippocampus. Muscarinic blockade weakens the positional firing of most place cells and therefore renders them less useful for precise representation of the environment. This effect may underlie the difficulties in spatial learning and problem solving caused by abnormalities of cholinergic transmission.

Action potentials and relations to the theta rhythm of medial septal neurons in vivo

The influence of the medial septal nucleus and the nucleus of the diagonal band of Broca (MS-DB) on the hippocampal theta rhythm includes both cholinergic and gamma-aminobutyric acid (GABAergic) components. To understand the intrinsic septal interactions and the separate contributions of the cholinergic and GABAergic septohippocampal neurons to the theta rhythm in behaving animals, it is essential to be able to identify these two classes from extracellular recordings. Here the durations of extracellularly recorded action potentials are compared with the other characteristics of the neurons. Extracellular recordings were taken from neurons of the MS-DB both in freely moving rats (114 cells) and in urethane-anesthetized rats (112 cells). These were compared with intracellular recordings taken from MS-DB neurons in urethane-anesthetized rats (58 cells). Hippocampal EEG was recorded from above the CA1 pyramidal cell layer (CAI theta) and near the hippocampal fissure (dentate theta) to compare the firing phase across cells. Here it is shown that two major types of rhythmically bursting cells in the MS-DB that had been distinguished previously in intracellular recordings in vivo are also separable in extracellular recordings in vivo on the basis of the durations of their action potentials. In both awake and anesthetized rats the main properties of the two cell types were found to differ: firing rate, phase-relation to the hippocampal theta rhythm and sensitivity of their rhythmicity to blockade of muscarinic transmission. As was previously shown for intracellular recordings in anesthetized rats, it is shown here that in awake rats, too, the more rapidly firing brief-spike (putative GABAergic) cells fired with highest probability on the negative phase of the dentate theta, whereas the more slowly firing long-spike (putative cholinergic) cells fired mostly on the positive phase. Previous work showed that in intracellular recordings from anesthetized rats the rhythmic firing of most brief-spike cells was still retained even during muscarinic blockade, but that of most long-spike cells was lost. Here we also report a recategorization according to spike duration of existing extracellular recordings taken from anesthetized rats, confirming the above observation with much larger numbers of cells. Three additional major new findings are also reported here. (1) In awake rats, muscarinic blockade has relatively little effect on either cell type. (2) Under anesthesia, the firing rates of both cell types are lower than in awake rats, but the effect is greater on the long-spike cells, where the anesthesia also reduces the rhythmicity of the cell firing. (3) Rhythmicity of the putative GABAergic cells is also retained after local injection of GABA-A antagonist, whereas that of the putative cholinergic cells is eliminated. We conclude that either systemic muscarinic blockade or urethane anesthesia alone have relatively little effect on neurons in the defined above MS-DB, but a combination of the two has profound effects on the rhythmicity of the cholinergic cells, largely sparing the GABA-ergic cells. Taken together, the results suggest that generation of theta rhythm requires a background of excitatory influences on the hippocampus (that can be maintained by either muscarinic or glutamatergic inputs) in combination with the phasic disinhibitory action mediated by the GABAergic MS-DB projection. They also provide additional support for the notion that the phasic activity in local collaterals of GABAergic MS-DB cells contributes to the phasic modulation of the firing of cholinergic septohippocampal neurons.

Intracellular recordings from medial septal neurons during hippocampal theta rhythm

The electrophysiological properties of neurons of the medial septal nucleus and the nucleus of the diagnonal band of Broca (MS/DB) were studied using intracellular methods in urethane-anesthetized rats. Three types of rhythmically bursting neurons were identified in vivo on the basis of their action potential shapes and durations, afterhyperpolarizations (AHPs), membrane characteristics, firing rates and sensitivities to the action of muscarinic antagonist: (1) Cells with short-duration action potentials and no AHPs (2 of 34 rhythmic cells, 6%) had high firing rates and extremely reliable bursts with 6-16 spikes per theta cycle, which were highly resistant to scopolamine action. (2) Cells with short-duration action potentials and short-duration AHPs (8 of 34 rhythmic cells, 24%) also had high firing rates and reliable bursts with 4-13 spikes per theta cycle, phase-locked to the negative peak of the dentate theta wave. Hyperpolarizing current injection revealed a brief membrane time constant, time-dependent membrane rectification and a burst of firing at the break. Depolarizing current steps produced high-frequency repetitive trains of action potentials without spike frequency adaptation. The action potential and membrane and characteristics of this cell type are consistent with those described for GABAergic septal neurons. Many of these neurons retained their theta-bursting pattern in the presence of muscarinic antagonist. (3) Cells with long-duration action potentials and long-duration AHPs (24 of 34 rhythmic cells, 70%) had low firing rates, and usually only 1-3 spikes per theta cycle, locked mainly to the positive peak of the dentate theta rhythm. Hyperpolarizing current injection revealed a long membrane time constant and a break potential; a depolarizing pulse caused a train of action potentials with pronounced spike frequency adaptation. The action potential and membrane properties of this cell type are consistent with those reported for cholinergic septal neurons. The theta-related rhythmicity of this cell type was abolished by muscarinic antagonists. The phasic inhibition of "cholinergic" MS/DB neurons by "GABAergic" MS/DB neurons, followed by a rebound of their firing, is proposed as a mechanism contributing to recruitment of the whole MS/DB neuronal population into the synchronized rhythmic bursting pattern of activity that underlies the occurrence of the hippocampal theta rhythm.

Modulation of the reaction of hippocampal neurons to sensory stimuli by cholinergic substances

The influences of increasing endogenous acetylcholine (eserine) and its blockade (scopolamine) on the effects of sensory stimuli were analyzed through the extracellular recording of the activity of individual hippocampal neurons of awake rabbits. An increase in the level of acetylcholine, accompanied by the appearance of stable theta rhythm, leads to a substantial decrease in the reactivity of neurons, the suppression, attenuation, and inversion of the majority of inhibitory reactions and of a substantial proportion of activational reactions including on-responses of a specific type. At the same time, a limited group of activational reactions is intensified and extended against the background of eserine. Scopolamine, which blocks theta rhythm, does not change or intensifies inhibitory and some activational reactions, including on-responses. Tonic reactions are shortened; however, their gradual extinction disappears. The effects described are preserved in the hippocampus in the presence of basal undercutting of the septum which eliminates ascending brainstem pathways. These data make it possible to draw the conclusion that, under normal conditions, a new (significant) sensory stimulus elicits in the hippocampus an initial stoppage (reset) of activity with the coordinated triggering of theta rhythm and the passage against this background of signals along the cortical input in a specific phase relationship to it. The period of theta modulation switched on by the signal fosters its recording and the limitation of the passage of subsequent, interfering signals. The septohippocampal influences may thus support the mechanism of selective attention, as a necessary precondition for memory.

Modulation of the influences of cortical input on hippocampal neurons by cholinergic substances

The cholinergic modulation of responses of individual neurons and of the focal potentials of the hippocampus, induced by electrical stimulation of the perforant path or mossy fibers were studied in two groups of unanesthetized rabbits, one with an intact septal region (IS), and one with its basal undercutting (BU). In all of the animals the responses to stimulation were blocked or markedly suppressed in a substantial portion of the neurons (50% in IS, 69% in BU) against the background of the administration of eserine. Facilitation of the responses was observed in 10 and 8% of cases, respectively. Scopolamine restored the initial reactivity of hippocampal neurons and intensified responses to stimulation of the perforant path. The effect of eserine was reproduced by stimulation of the medial septal region (MS-DB). The depressive effect of stimulation of the MS-DB was intensified by the administration of eserine and blocked by scopolamine. Brief conditioning stimulation of the MS-DB which imitates a theta salvo facilitated responses to test stimulation of the MS-DB with delays of 70-150 msec, but suppressed them at smaller and greater intervals. Focal potentials in response to stimulation of the perforant path in CA1 were suppressed to an equal extent (by 43%) during sensory stimulation inducing natural theta rhythm, during the action of eserine, and with stimulation of the MS-DB In the BU group, these effects led to the complete suppression of focal potentials; scopolamine restored them. It is hypothesized that the principal function of the septohippocampal cholinergic input resided in the negative filtration of signals arriving against the background of theta rhythm that has been turned on by another, preceding influence, as a result of which their interference with the processing and recording of received information is prevented.

Modulation of septal influences on hippocampal neurons by cholinergic substances

The effects of electrical stimulation of the medial septal area (MS-DB) for the purpose of distinguishing and assessing the cholinergic component of the septohippocampal input were investigated in awake rabbits in chronic experiments. Initial inhibitory effects of a standard duration of 40-140 msec (54%) predominated in the intact rabbits. In animals with chronic basal undercutting of the MS-DB, initial inhibitory reactions predominated absolutely (90%). An increase in the level of endogenous acetylcholine by administration of eserine led to a partial or complete suppression of all effects of stimulation in 78% of the hippocampal neurons of the intact rabbits against the background of intensification of the theta modulation of the activity of hippocampal neurons. Scopolamine removed theta modulation and restored the reactivity of neurons to stimulation of the MS-DB. These influences of cholinergic substances were maintained in the animals with basal undercutting of the MS-DB. It is inferred that the general initial influence of septal input on neurons of the hippocampus is expressed in the suppression of their activity ("reset"), which depends on the noncholinergic (GABAergic) component of the septohippocampal connections. The cholinergic component limits the effectiveness of both extraseptal (brainstem) and primary inhibitory septal influences on hippocampal neurons.

[The modulation by cholinergic substances of the neuronal reaction of the hippocampus to sensory stimuli]

Neuronal activity was recorded extracellularly in the hippocampus of waking rabbits. Modifications of the effects of sensory stimulation were analysed on the background of increased level of endogenous acetylcholine (injection of physostigmine) and during its blocking by scopolamine. Significant decrease of responsiveness (about 40%) of the hippocampal neurons to sensory stimuli occurred after physostigmine injection. Suppression, decrease and reversal of the inhibitory responses (including initial reset phase) and of some excitatory reactions, including on-effects was observed on the background of stable theta-rhythm. However, a limited group of excitatory responses was augmented and prolonged by physostigmine. Under scopolamine action the responsiveness of the neurons was not changed. Some of the inhibitory and excitatory effects, especially on-responses were augmented. Tonic responses became shorter, but they were stably reproduced without the typical gradual habituation. All the effects were also present in the hippocampus after the basal septal undercutting eliminating ascending brainstem input. It is suggested that under the normal conditions a new or significant sensory stimulus evokes in the hippocampus initial inhibitory reset of neuronal activity with the following coordinated triggering of theta-rhythm and arrival of the cortical input signal phase-locked to it. During the period of theta triggered by a stimulus its processing and fixation in the memory occurs, while the other, interfering stimuli are actively filtered out. Thus, the septo-hippocampal interactions provide for mechanisms of selective attention as a necessary condition of memory trace formation.

[The modulation by cholinergic substances of the influences of cortical input on hippocampal neurons]

Cholinergic modulation of the single-cell responses and field potentials evoked in the hippocampus by electrical stimulation of its cortical input was investigated in two groups of chronic anaesthetized rabbits--with the intact septum (IS) and basally undercut (BS). In both groups of animals responses to stimulation of the perforant path (PP) or mossy fibers were blocked or significantly suppressed in substantial proportion of the neurons (50% in IS, 69% in BS) by i.v. physostigmine injection; facilitation of responses was observed in minor groups of the neurons (10% and 8%). Scopolamine restored initial responsiveness of the hippocampal neurons and augmented responses to stimulation in some of them (37.5% in IS, 65% in BS). Effect of physostigmine was reproduced by single stimuli applied to the MS--DB. Field potentials evoked by PP stimulation in CA of IS group were equally suppressed (by 43%) by the sensory stimulation evoking natural theta, by physostigmine and MS--DB stimulation. In BS group of animals these influences completely suppressed the focal potentials to PP stimulation. Scopolamine restored the focal potentials. It is concluded that the main function of the septo-hippocampal cholinergic input consists in filtering out the signals appearing at the background of the theta-rhythm triggered by a previous signal thus preventing their interference with its processing and recording.

[The modulation of septal influences on the hippocampal neurons by cholinergic substances]

To evaluate the functional significance of the cholinergic component of the septo-hippocampal input the effects of electrical stimulation of the medial septal area (MS--DB) on the hippocampal neurons were studied in chronic waking rabbits. Initial inhibitory effects 40-140 msec long were observed in 54% of neurons in intact rabbits. Other neurons usually responded by a diffuse excitation. Initial inhibitory effects absolutely dominated (90%) in the animals with the septum basally undercut. Increase of the endogenous acetylcholine level by physostigmine completely or partially blocked the effects of the stimulation in 78% of the hippocampal neurons at the background of increased theta-modulation. Scopolamine restored responsiveness of the neurons to the MS--DB stimulation. These effects were preserved in the animals with the septal undercutting. It is concluded that the general effect of the MS--DB on the hippocampal neurons consists in a non--cholinergic (GABA-ergic) primary suppression (reset) of their activity. The cholinergic component limits the efficacy of both extraseptal (brainstem) and septal inhibitory influences.

The spontaneous activity of hippocampal neurons during the modulation of theta rhythm by cholinergic substances

A statistical analysis of the baseline activity of neurons, recorded intracellularly in the hippocampus of awake, nonimmobilized rabbits in three states, control and during the systemic administration of eserine and scopolamine, was carried out. Neurons of the hippocampus were additionally tested in a similar manner following the chronic basal undercutting of the septum, removing stem influences. The cholinergic substances regulate the number of neurons of the hippocampus having theta modulation and the degree of its stability, but do not influence its frequency. When the cholinergic theta rhythm is activated, regularization of the activity takes place with the suppression of delta modulation and of "complex spikes"; its blockade is accompanied by the opposite changes. Both substances stably alter the level of the baseline frequency of discharges of the majority of neurons, although the total average frequency remains constant. Regression analysis shows the predominance of a decrease in the activity in high-frequency (> 25 spikes/sec) and an increase in the low-frequency (< 25 spikes/sec) neurons during the effect of both substances. The constancy of the total average frequency and the unidirectionality of the shifts in the level of discharges of the neurons during the intensification (eserine) and blockade (scopolamine) of the cholinergic component of the theta rhythm points to the fact that the cholinergic septal input directly influences mainly the structure but not the level of the activity of the hippocampal neurons.

Theta modulation of neurons of the hippocampus of the rabbit and its interrelationship with other parameters of spontaneous and evoked activity

The reliability of the existing functional criteria of the differentiation of pyramidal ("neurons with complex spikes") and inhibitory ("theta neurons") cells of the hippocampus is examined on the basis of a statistical analysis of the spontaneous and evoked activity of neurons of the hippocampus of the awake rabbit. The analysis shows that the parameters of average frequency, the presence of theta modulation of activity, the behavior of the neurons in situations evoking theta rhythm in the EEG of the hippocampus (inhibition or activation during the effect of sensory stimuli), and the character of the influences of stimulation of the medial septal region of the internal connections of the hippocampus do not permit the reliable identification of different types of neurons of the hippocampus in the awake rabbit. The available data on the functional classification of neurons of the hippocampus are discussed in connection with notions regarding their state in situations associated with the generation of theta rhythm.

Acetylcholine, theta-rhythm and activity of hippocampal neurons in the rabbit--II. Septal input

The aim of this paper was to evaluate the cholinergic component of the septohippocampal input signals in neuronal activity of the hippocampal fields CA1 and CA3 recorded extracellularly in chronic alert rabbits. Effects of electrical stimulation of the medial septal area were analysed in the control state, on the background of an increased level of endogenous acetylcholine (by physostigmine injection) and during its blockade by antimuscarinic drugs (scopolamine, atropine). Two groups of animals were used in the experiments: intact rabbits and rabbits with complete chronic undercutting of the septum, depriving the septohippocampal system of ascending medial forebrain bundle afferents. Primary inhibitory effects of standard duration (40-140 ms) evoked by medial septal area stimulation dominated in the hippocampus of intact rabbits (54%), though some neurons responded by initial diffuse excitation (37.5%); responses by single-spike on-effects were observed in a minority of neurons (8.5%). The primary suppression of activity prevailed (90%) in animals with basal undercutting of the septum. In intact rabbits under physostigmine action, the effects of medial septal area stimulation were depressed or completely blocked in 78% of hippocampal neurons on the background of increased theta modulation of activity. Neuronal responses to medial septal area stimulation recovered at the background of muscarinic antagonists. These effects of cholinergic drugs were reproduced in animals without medial forebrain bundle. It is concluded that the initial effect of the septal input upon the hippocampal neurons consists of a general suppression of their activity (reset), depending upon a non-cholinergic (presumably GABAergic) component of the septohippocampal connections.(ABSTRACT TRUNCATED AT 250 WORDS)

Acetylcholine, theta-rhythm and activity of hippocampal neurons in the rabbit--IV. Sensory stimulation

Modifications of responses of hippocampal neurons to sensory stimuli at the background of increased endogenous acetylcholine level (injection of physostigmine) and during blocking by scopolamine were analysed in the chronic alert rabbit. A significant decrease of reactivity (about 40%) of hippocampal neurons to sensory stimuli occurred after physostigmine injection, inducing stable theta modulation. Suppression and decrease of inhibitory responses (including initial reset phase) and of some excitatory reactions (including on-effects) were observed. However, a limited group of excitatory responses was augmented and prolonged under physostigmine action. Scopolamine, which blocked electroencephalogram theta-rhythm, did not change the responsiveness of hippocampal neurons. Some of the inhibitory and excitatory effects of sensory stimuli, especially on-responses, were strongly facilitated. Tonic responses were shorter, but they were stably reproduced without typical gradual habituation. All these effects were also present in the hippocampus after basal undercutting of the septum, which eliminates ascending brainstem input. It is suggested that under normal conditions a new or significant sensory stimulus evokes, in the hippocampus, an initial inhibitory reset of neuronal activity with subsequent coordinated triggering of rhythmic theta modulation by the septal input and arrival of the cortical input signal phase-locked to it. During the period of theta triggered by the stimulus, its processing and fixation in memory occurs, while the other, interfering stimuli, which are not phase-locked to the ongoing theta activity, are actively filtered out. Thus, septohippocampal interactions may participate in the organization of selective attention as a necessary condition for memory trace formation.

Acetylcholine, theta-rhythm and activity of hippocampal neurons in the rabbit--I. Spontaneous activity

The background activity of hippocampal neurons was recorded extracellularly in waking rabbits in the control state and after systemic injections of physostigmine and scopolamine. Similar analysis was done in the hippocampus chronically deprived of ascending brainstem afferents. Cholinergic drugs control the number of hippocampal neurons with theta modulation and stability, but not the frequency of theta modulation. Increase of endogenous acetylcholine also resulted in regularization of the activity with suppression of delta modulation and complex spike discharges; its blockade produced the opposite changes. Both drugs changed the level of background activity in the majority of the neurons, but the overall mean frequency did not vary between the states. Regression analysis demonstrated significant negative correlations with dominating decrease in the level of discharges in high-frequency neurons (> 25 spikes/s) and its increase in low-frequency ones (< 25 spikes/s) after injection of both drugs. Stability of the overall mean frequency and uniformity of its shifts during both stimulation and suppression of the cholinergic component of theta-rhythm presumably indicate that the frequency of background activity, unlike its pattern, is not directly controlled by the cholinergic septal input.

Acetylcholine, theta-rhythm and activity of hippocampal neurons in the rabbit--III. Cortical input

Cholinergic modulation of single cell responses and field potentials evoked in the hippocampus by electrical stimulation of the perforant path and mossy fibres was investigated in two groups of chronic unanesthetized rabbits--with intact hippocampus and with basally undercut septum (without ascending medial forebrain bundle afferents). In both groups of animals responses to stimulation were blocked or significantly depressed by i.v. physostigmine injection in many neurons (50% in the intact hippocampus and 69% in the hippocampus without medial forebrain bundle). In minor groups of neurons (10 and 8%, respectively), facilitation of responses was observed. Scopolamine restored initial responsiveness of hippocampal neurons and augmented effects of stimulation in some of them. The effect of physostigmine was reproduced by stimulation of the medial septum. Depressive influence of medial septal area stimulation was increased by physostigmine and blocked by scopolamine. Population spikes evoked by stimulation of the perforant path of the intact group were equally suppressed (by 43%) during sensory stimulation evoking natural theta, after physostigmine and after medial septal area stimulation. In the group of animals without medial forebrain bundle these influences resulted in a complete suppression of field potentials; scopolamine restored them. It is concluded that the main function of the septohippocampal cholinergic input consists of filtering out the signals appearing at the background of theta-rhythm triggered by a previous signal, thus preventing their interference with its processing and registration.

[The spontaneous neuronal activity of the hippocampus during the modulation of the theta rhythm by cholinergic substances]

Background activity of the hippocampal neurons, extracellularly recorded in waking chronic rabbits, was analysed in control state and after systemic injection of physostigmine and scopolamine. Similar analysis was done in the hippocampus chronically deprived of ascending brain stem afferents. Cholinergic drugs controlled the number of hippocampal neurons with theta-modulation and the degree of its stability but not the frequency. Activation of cholinergic theta-rhythm resulted also in regularization of activity with suppression of delta-modulation and complex spike discharges; its blockade was accompanied by the opposite changes. Both drugs shifted the level of background activity in the majority of neurons, but the overall mean frequency did not vary between the states. Regression analysis demonstrated significant negative correlations with dominating decrease in the level of activity in high-frequency neurons ( > 25 sp/s) and its increase in low-frequency ones ( < 25 sp/s) after injection of both drugs. Stability of the overall mean frequency and uniformity of its shifts presumably indicate that the frequency, unlike the pattern of the background activity, is not directly controlled by the cholinergic septal input.

[The theta modulation of rabbit hippocampal neurons and its correlation with other indices of spontaneous and evoked activity]

Reliability of the existing functional criteria for differentiation of pyramidal ("complex spike neurones") and inhibitory ("theta neurones") cells in the hippocampus of waking rabbit is evaluated on the basis of statistical analysis of neuronal spontaneous and evoked activity. The analysis shows, that the criteria of mean frequency, presence of theta modulation, neuronal behaviour in situations provoking EEG theta rhythm (e.g., excitation or inhibition during presentation of sensory stimuli), effects of medial septum and intrahippocampal stimulation do not permit reliable identification of the hippocampal neuronal types in the waking rabbit. The data on functional classification of the hippocampal neurones are discussed in connection with existing suggestions about their state in situations inducing theta rhythm generation.

[The role of GABA-ergic regulation in organizing spontaneous and evoked activity of the septal neurons]

Effects of GABA, pentobarbital and picrotoxin upon spontaneous and evoked activity of neurones of the medial septal nucleus and the nucleus of the diagonal band (MS-DB) were investigated in the guinea pig septal slices. GABA and pentobarbital have similar effect upon all neurones, but the cells with a regular single spike and rhythmic burst activity of pacemaker type were less sensitive to their inhibitory influence. Picrotoxin affects neither frequency, nor pattern of activity. Electrical stimulation of the medial forebrain bundle evoked initial suppression of activity in majority of the neurones (74%); the remaining cells reacted mainly with an initial burst. GABA and pentobarbital increased the duration of the initial inhibition and revealed it in all cells with initial excitation in the control state. Picrotoxin did not influence this type of response, but revealed initial short-latency bursts in the cells with inhibitory effect in control state. The experiments show double nature of the effect of afferent stimulation controlling the activity of the MS-DB neurones. The mechanism of synchronization of the rhythmic activity in MS-DB, resulting in generation of the hippocampal theta-rhythm, is discussed.

[Modulation of signal transmission by the septal neurons of the rabbit during action on the cholinergic system]

The neurons of the medial septum-diagonal band complex (MS-DB) were examined extracellularly in chronic unanaesthetized rabbits. Low-frequency (3-21 Hz) electrical stimulation was applied to the medial forebrain bundle within the horizontal limb of DB. In a group of the MS-DB units with stable background theta bursts the typical response consisting of entrainment of the phase-locked theta cycles was changed neither by physostigmine, nor by cholinergic-blocking drugs (scopolamine and atropine). In major group of the MS-DB units (60%), physostigmine completely blocked effects of electrical stimulation. This occurred both in the units with entrainment of the theta cycles and in the units with the effects of primary suppression of activity or (to a lesser extent) with single-spike discharges following the repetitive stimuli up to high frequencies. The cholinergic-blocking drugs restored and sometimes increased the initial reactivity of the MS-DB units. It is suggested, that intraseptal cholinergic system exerts a powerful gating effect upon transmission of signals to the hippocampus, probably, by the mechanisms of presynaptic inhibition.

Modulation of the afferent input to the septal neurons by cholinergic drugs

The effects of cholinergic drugs upon the evoked activity of extracellularly recorded neurons of the medial septal nucleus-nucleus of the diagonal band (MS-DB complex) were tested in unanesthetized rabbits. Electrical stimulation of MFB resulted in entrainment of the background theta-cycles in the neurons with strong rhythmic discharge (types I and II). Phase-locking of the background theta-cycles to the stimulus occurred 'by the burst', or 'by the pause' within the theta-range of frequencies (3-12 Hz). Single-spike responses, following up to 30 Hz and more, were also evoked by MFB stimulation, especially in the cells with weak theta-modulation (type III) or without it (type IV). Injection of physostigmine increased background theta-modulation of neuronal activity and simultaneously blocked or diminished responses to repetitive MFB stimulation in 82% of the MS-DB units, independent of their type of response. Driving of theta-cycles both 'by the burst' and 'by the pause' was ineffective or drastically reduced. Single spike responses disappeared or became unstable, though their minimal latencies did not change. Initial inhibitory responses were blocked or became significantly shorter. Antimuscarinic drugs, scopolamine and atropine, which abolished theta-modulation in many MS-DB units, restored responses and sometimes enhanced them. Repetitive stimulation of the MFB in this condition was effective up to the high frequencies, well beyond the theta-range. Thus, the majority of the MS-DB units did not respond to the afferent stimuli during prominent theta-activity evoked by physostigmine. The role of the septal cholinergic system in gating of afferent input during the theta-state and its importance for learning and memory is suggested.

[Frequency modulation of theta bursts of the septal neurons in the removal of ascending afferent effects in rabbits]

Medial septal-diagonal band (MS-DB) units were examined extracellularly in chronic rabbits under two experimental conditions: 1) in an intact septum, under anaesthetic doses of pentobarbital (40 mg/kg, i.v.); 2) in a basally undercut septum of unanaesthetized rabbits. The background rhythmic burst activity was undistinguishable in both states. Low-frequency electric stimulation of afferent inputs (MFB, CA1, LS) led to entrainment of the theta-cycles. The upper limit of following was almost normal in the undercut septum, but was strongly reduced under pentobarbital. In units with the driving "by pause", the duration of the initial silent period under pentobarbital was increased almost twofold but in the basally undercut septum it was the same as in the normal state. Some MS-DB units with weak or absent theta-modulation reacted to stimulation by stimulus-locked single spike discharges which followed up to high frequencies in both conditions.

[Effect of anticholinergic drugs alone and in combination with nembutal on burst theta-neurons of the rabbit septum]

Activity of neurones with rhythmic theta-bursts was recorded in the medial septum--diagonal band complex of the waking rabbits with intact and deafferented septum. Effects of anticholinergic (scopolamine, atropine) and cholinomimetic (physostigmine) drugs were investigated after i.v. injection. Cholinoblocking drugs in doses, suppressing the theta-rhythm in the hippocampal EEG, eliminated rhythmic activity in some cells with weak theta-modulation. Theta-bursts persisted in cells with stable continuous rhythmicity, though its regularity decreased in some of them. Strong reticular or sensory stimulation evoked an increase of burst frequency, involvement of additional septal cells into rhythmic activity and appearance of the theta-rhythm in the hippocampal EEG. Neither anticholinergic, nor cholinomimetic drugs influenced the frequency and basic characteristics of theta-bursts in any condition tested. The anticholinergic drugs have no selective effect upon low-frequency theta-bursts. The septohippocampal connections contain a significant non-cholinergic component. The theoretical concept of the septum as a sole source of the whole frequency band of the theta-rhythm is proposed.

Control of the neuronal rhythmic bursts in the septal pacemaker of theta-rhythm: effects of anaesthetic and anticholinergic drugs

The drugs, described as blocking the high-frequency (pentobarbital) or low-frequency (scopolamine, atropine) theta rhythm of the hippocampal electroencephalogram, were tested upon the rhythmically bursting septal cells. Three groups of chronic, unanaesthetized rabbits were used for the experiments: with intact septum; with septohippocampal disconnection; with complete basal undercutting of the septum, depriving it of ascending brainstem influences (MFB lesion). While the frequency and other parameters of theta bursts did not differ in the first two groups (5.2-5.5 Hz), in MFB-lesioned septum their frequency was significantly lower (3.5 Hz). Intravenous injection of pentobarbital suppressed theta bursts in some cells with unstable, periodic rhythmic activity and lowered the frequency of the bursts in continuously bursting cells. The parameters of bursts in intact and hippocampectomized septum under pentobarbital did not differ from those of undercut septum in undrugged state. Acetylcholine-blocking drugs suppressed theta modulation in some intermittently bursting cells, but only slightly decreased regularity of the bursts in some cells with continuous theta bursting even in sublethal doses; physostigmine has the opposite effect. Neither scopolamine and atropine, nor physostigmine influenced frequency of theta bursts in any way. Sensory or reticular stimulation could temporarily restore both the theta rhythm of hippocampal EEG and the rhythmic bursting of some septal cells under pentobarbital or anticholinergic drugs. On the basis of the experiments a unitary concept of theta rhythm origin is proposed. Pentobarbital influences ascending excitatory input to the septum, which results in a decrease of the burst frequency in the limited group of septal cells, regarded as endogenous bursting pacemakers, and in restriction of the population of high-threshold secondary rhythmic cells, synaptically involved in the rhythmic process. Anticholinergic drugs do not influence the pacemaker cells, but block intraseptal and septohippocampal cholinergic transmission. Both cholinergic and non-cholinergic neurons projecting to the hippocampus exist among septal cells synaptically involved in the rhythmic activity.

[Comparative characteristics of septal "burst" neurons after elimination of afferent effects of the reticular formation in the rabbit]

Comparative analysis of characteristics of rhythmic theta-activity in the neurones of the medial septal nucleus and nucleus of diagonal band was performed in intact rabbits after. i. v. injection of pentobarbital, and in rabbits with chronic lesion of the ascending brain-stem afferent fibers. In both conditions theta-bursts disappeared in some cells with unstable periodic rhythmic modulation; substantial population of the septal units preserved regular burst activity. Main characteristics of theta-bursts were almost identical in both states, their mean frequency decreased to 3.5 Hz. The theta-rhythm in hippocampal EEG was usually absent; but low-frequency rhythmic activity could be evoked by electrical or sensory stimulation as well as by injection of bemegrid or physostigmine. The data show that the ascending brain-stem afferents control: the frequency of the bursts in a population of septal units regarded as bursting pace-maker cells; the total number of the septal cells secondarily (synaptically) involved into rhythmic activity. The effect of pentobarbital upon theta-rhythm results from elimination of these influences upon the septal cells.

[Frequency modulation of theta volleys of septal neurons during rhythmic oligosynaptic stimulation in the rabbit]

Activity of the neurones with stable theta-bursts was recorded extracellularly in intact and hippocampectomized septum of unanaesthetized chronic rabbits during low-frequency (3-17 Hz) stimulation of horizontal limb of the diagonal band or the lateral septal nucleus. Gradual entrainment and phase-locking of the spontaneous theta-cycles occurred. Two types of entrainment were observed: "entrainment by pause", where interburst interval was reset by the stimuli; and "entrainment by burst", where bursts were time-locked to the stimuli. Such reorganization of the spontaneous bursts occurred in a narrow frequency range of stimulation (from 4 Hz up to 9-12 Hz), with the best resonance following in the range of "basic" theta frequencies of the awake rabbit (5-6 Hz). With stimulation beyond the theta-range three phenomena occurred: shift of the burst frequencies to higher or lower harmonics of stimulation frequencies; complex interactions of basic background frequency with the rhythm of stimulation ("beating"); escape from the influence of the stimuli with return to background theta-burst frequency.

Frequency modulation of neuronal theta-bursts in rabbit's septum by low-frequency repetitive stimulation of the afferent pathways

Activity of the neurons with stable theta-bursts was recorded extracellularly in intact and hippocampectomized septum of unanesthetized chronic rabbits during low-frequency (3-17 Hz) stimulation of horizontal limb of diagonal band or lateral septal nucleus. The stimulation rarely evoked standard oligosynaptic single-spike responses in stable theta-bursting units. Instead, gradual entrainment and phase-locking of the spontaneous theta-cycles occurred. Two types of entrainment were observed: "entrainment by the pause", where interburst interval was reset by the stimuli; and "entrainment by the burst", where bursts were time-locked to the stimuli. Such reorganization of the spontaneous bursts occurred in the narrow frequency range of stimulation (from 4 Hz up to 9-12 Hz), with the best resonance following in the range of "basic" theta frequencies of the awake rabbit (5-6 Hz). With stimulation beyond the theta-range three phenomena occurred: shift of the burst frequencies to higher or lower harmonics of stimulation frequencies; complex interactions of basic background frequency with rhythm of stimulation ("beating"); return to background theta-burst frequency in spite of continuing stimulation ("escape"). The properties of the septal theta-bursting cells as presumable intrinsic bursting pacemakers are discussed.

[Regulation of the theta activity of septal neurons by cortical and brain stem structures]

Neuronal rhythmic burst activity was recorded extracellularly in the intact septum (34 cells) and in septum deprived of hippocampal connections (100 cells) during electrical stimulation of various neocortical areas, midbrain and pontine reticular formation, and medial forebrain bundle (MFB) in the preoptic area. All structures tested influenced burst production in septal cells. Cortical influences might phasically trigger or regularize theta-activity, but more often they suppressed and desynchronized it. The reticular stimulation tonically increased frequency, regularity, and density of the theta-bursts. MFB influences were similar, but less effective. Increasing the intensity of the stimulating current resulted in a logarithmic increase of theta-burst frequency up to the upper limit of the EEG theta-frequencies (9-10 Hz). Short electrical or sensory stimulation reset the burst activity so that the series of bursts followed each other for several seconds in a stable pattern, which was reproduced by each next application of the same stimulus. Regulation of the theta-rhythm by direct influences upon various types of septal units without hippocampal involvement is discussed.

[Activity of hippocampal neurons deprived of ascending brain stem connections]

97 units of the hippocampal fields CA3 and CA1 were recorded extracellularly in chronic unanaesthetized rabbits after complete basal undercutting of the septum. In activity of about one third of the units slow rhythmic bursts (3,3-4,5 Hz) were present. Low frequency theta-like rhythm was present in EEG of the hippocampus. Reactivity to sensory stimuli was unusually low (46-47% of reactive units). Specific and phasic effects of stimuli, typical of the normal field CA1, were observed in both fields. The majority of the reactive neurons respond to sensory stimuli by prolonged shift of the mean frequency of discharges, by regularization of the rhythmic component, or by gradual increase of diffuse activation. Effects of suppression of activity by sensory stimuli were very rare. The data are discussed in the light of presumed difference of reinnervation by axonal sprouting in conditions of basal undercutting of the septum and complete septo-hippocampal disconnection.

Neuronal activity of the septum following various types of deafferentation

Extracellular recording of neuronal activity of septal nuclei was performed in unanesthetized chronic rabbits in the following conditions: (1) intact septal nuclei; (2) complete interruption of connections with hippocampus; (3) complete basal undercutting of septum. Additional series of records were performed in transversal guinea pig septal slices incubated in vitro. Computer analysis of activity was used. Interruption of hippocampal connections increased the mean frequency of discharges in the lateral septal nucleus 2-fold. Theta-bursts were preserved and the proportion of theta-bursting neurons increased from 28 to 42%. The mean frequency of theta-bursts was slightly augmented (mean 5.4 cps versus 5.2 cps in intact septum). Twenty-six per cent of neurons were characterized by regular, pacemaker-like discharges, while in the normal septum such units were almost completely absent. After septal undercutting the mean frequency of spontaneous activity was increased 2-fold in the medial septal nucleus. Rhythmic bursts were again preserved, the proportion of bursting neurons was close to normal (25%), but the mean frequency of bursts was significantly lower (mean 3.3 cps). Rhythmic activity was extremely regular. Neurons (21%) with pacemaker-like activity were also present. In septal slices neurons possessed relatively high spontaneous activity (mean 10.2/sec). In the majority of units (64%) activity was highly regular. Twenty-two per cent of the units possessed rhythmic bursts in their activity with frequencies of 2.0-5.1 cps. The role of intrinsic septal mechanisms and afferent influences in generation and modulation of theta-bursts is discussed.

[Effect of complete basal undercutting of the septum on the activity of its neurons]

Spontaneous and evoked by sensory stimuli neuronal activity of medial and lateral septal nuclei deprived of basal connections, was investigated in chronic unanaesthetized rabbits. After basal undercutting of septum many neurones in both nuclei produced highly regular "pacemaker-like" activity. The proportion of neurones with rhythmic bursts of spontaneous activity did not change, but the frequency of bursts decreased to 2.8--3.9 per second. Reactivity of neurones to sensory stimuli was low in both nuclei. Electric stimulation of the hippocampus was highly effective. The majority of septal neurones responded to each stimulus by a short (30--150 msec) inhibition. The problems in connection with possible organization of rhythmic theta-activity of septal neurones are discussed.

[Neuronal activity of the "isolated hippocampus"]

Spontaneous and evoked activity of the field CA1 neurones were investigated in long-lasting experiments on unanesthetized rabbits with "isolated hippocampus" (destruction of both septal and perforant path's afferents). Testing by various sensory stimuli as well as by electrical stimulation of Schaffer's collateral's, subiculum, posterior cingulate cortex, neocortex and midbrain reticular formation was used. The sensory reactions were virtually absent during first 1--2 weeks after the operation, but by the 4--6th week the level of reactivity became normal (65--70%). The first sensory reactions encountered were of diffuse tonic type; later on complex patterned on-responses appeared. The sources of hippocampal reactions, the possibility of plastic rearrangement of synapses and the absence of intrinsinc ability for generation of the theta-activity are discussed.

[Theta-bursts of hippocampal and septal neurons]

Investigation of neuronal activity on non-anaesthetized rabbits in non-stress chronic conditions revealed in the hippocampus a small proportion of neurones (10-12 percent) with a weak rhythmic (theta) modulation of their activity. In the medial and lateral septal nuclei this activity was more organized and was present in about one third of neuronal population. Disruption of septo-hippocampal connections completely abolished theta-rhythm in EEG and in neuronal activity of the hippocampus. Transection of the perforant path was followed by appearance of pronounced theta-bursts in a large group of neurones (38 percent in CA1 and 43 percent in CA3). Electrical stimulation of the hippocampus suppressed theta-bursts and decreased their frequency in the lateral spetal nucleus. Disruption of hippocampo-septal connections was followed by an increase of the population of septal neurones with theta-bursts (48 percent), which became more prounced. Some theoretical implications from these data concerning the nature of hippocampal theta-activity are discussed.

[Effect of disconnecting cortical input on the activity of hippocampal neurons]

Extracellular investigation of the CA1 and CA3 hippocampal neurones was performed in rabbits after complete bilateral transection of the perforant path (PP). Spontaneous activity was normal, but in a large group of neurones (38 per cent in CA1 and 43 per cent in CA3) rhythmic theta-bursts were observed, while in the normal state such neurones constituted only 9--12 per cent. The reactivity to sensory stimuli was enhanced, as well as the number of multimodal units. In both fields the majority of sensory reactions was of the tonic type; the number of phasic and "specific" effects, which are typical of the normal CA1 neurons, was strongly reduced. As a result, the CA1 field without PP closely resembled the normal field CA3. The phenomenon of gradual habituation of reactions was virtually absent; instead of that increase of reactions duration was obvious during repeated presentations of stimuli. The neurons were highly reactive to electrical stimulation of midbrain RF and displayed high correlation of characteristics of their reactions to sensory and reticular stimuli.

[Acetylcholinesterase in the rabbit hippocampal formation after destruction of the septohippocampal connections]

After deafferentation of the septofimbriate introitus to the rabbit hippocamp acetylcholinesterase activity lowers in homogenates and usbcellular fractions within different areas of the hippocampal formation and subiculum. The greatest decrease is observed in homogenates of the denticulate fascia (up to 25% of the norm). In the subcellular fractions the greatest decrease is observed in the activity of the coarse mitochondrial and MHN-20 fractions as compared to the nuclear and microsomal ones. When separating the coarse mitochondrial fraction within the linear gradient of sucrose density (0.7--1.6 M), the greatest decrease in acetylcholinesterase activity is observed in the upper layers of the gradient. Distruction of septohippocampal relations produces a strong drop in the activity of butyryl cholinesterase and lactate dehydrogenase in the fimbria homogenates.

[The effect of disconnecting the hippocampus from the septum on the activity of septal neurons]

Extracellular recording of neuronal activity was performed in the medial and lateral septal nuclei (MS and LS) in unanaesthetized rabbits after coagulation of septo-hippocampal connections. The MS neuronal activity had many pathological features. The LS activity was normal in every respect. Spontaneous activity, reactivity to sensory stimuli and main characteristics of responses to sensory stimuli were preserved in LS (and in a part of MS neurones). Sensory effects were augmented in intensity and duration, the number of neurones in LS with theta-bursts increased twofold, theta-bursts were more regular, than in control animals. These effects may be explained by an increase of ascending RF influences, which is supported by the fact of outstanding similarity between sensory and reticular effects in septal neurones after hippocampal disconnection. The number of units with inhibition of activity in response to sensory stimuli decreased, habituation of responses was absent. That means that hippocampal influences are necessary for the organization of inhibitory phenomena in the septum, and, above all, for processes of gradual habituation.