On the cellular and network bases of epileptic seizures

The highly interconnected networks of the mammalian forebrain can generate a wide variety of synchronized activities, including those underlying epileptic seizures, which often appear as a transformation of otherwise normal brain rhythms. The cerebral cortex and hippocampus are particularly prone to the generation of the large, synchronized bursts of activity underlying many forms of seizures owing to strong recurrent excitatory connections, the presence of intrinsically burst-generating neurons, ephaptic interactions among closely spaced neurons, and synaptic plasticity. The simplest form of epileptiform activity in these structures is the interictal spike, a synchronized burst of action potentials generated by recurrent excitation, followed by a period of hyperpolarization, in a localized pool of pyramidal neurons. Seizures can also be generated in response to a loss of balance between excitatory and inhibitory influences and can take the form of either tonic depolarizations or repetitive, rhythmic burst discharges, either as clonic or spike-wave activity, again mediated both by intrinsic membrane properties and synaptic interactions. The interaction of the cerebral cortex and the thalamus, in conjunction with intrathalamic communication, can also generate spike waves similar to those occurring during human absence seizure discharges. Although epileptic syndromes and their causes are diverse, the cellular mechanisms of seizure generation appear to fall into only two categories: rhythmic or tonic "runaway" excitation or the synchronized and rhythmic interplay between excitatory and inhibitory neurons and membrane conductances.

The neuronal basis for consciousness

Attempting to understand how the brain, as a whole, might be organized seems, for the first time, to be a serious topic of inquiry. One aspect of its neuronal organization that seems particularly central to global function is the rich thalamocortical interconnectivity, and most particularly the reciprocal nature of the thalamocortical neuronal loop function. Moreover, the interaction between the specific and non-specific thalamic loops suggests that rather than a gate into the brain, the thalamus represents a hub from which any site in the cortex can communicate with any other such site or sites. The goal of this paper is to explore the basic assumption that large-scale, temporal coincidence of specific and non-specific thalamic activity generates the functional states that characterize human cognition.

Control of spatiotemporal coherence of a thalamic oscillation by corticothalamic feedback

The mammalian thalamus is the gateway to the cortex for most sensory modalities. Nearly all thalamic nuclei also receive massive feedback projections from the cortical region to which they project. In this study, the spatiotemporal properties of synchronized thalamic spindle oscillations (7 to 14 hertz) were investigated in barbiturate-anesthetized cats, before and after removal of the cortex. After complete ipsilateral decortication, the long-range synchronization of thalamic spindles in the intact cortex hemisphere changed into disorganized patterns with low spatiotemporal coherence. Local thalamic synchrony was still present, as demonstrated by dual intracellular recordings from nearby neurons. In the cortex, synchrony was insensitive to the disruption of horizontal intracortical connections. These results indicate that the global coherence of thalamic oscillations is determined by corticothalamic projections.

Mechanisms underlying the synchronizing action of corticothalamic feedback through inhibition of thalamic relay cells

Early studies have shown that spindle oscillations are generated in the thalamus and are synchronized over wide cortical territories. More recent experiments have shown that this large-scale synchrony depends on the integrity of corticothalamic feedback. Previously proposed mechanisms emphasized exclusively intrathalamic mechanisms to generate the synchrony of these oscillations. In the present paper, we propose a cellular mechanism in which the synchrony is dependent of a mutual interaction between cortex and thalamus. This cellular mechanism is tested by computational models consisting of pyramidal cells, interneurons, thalamic reticular (RE) and thalamocortical (TC) relay cells, on the basis of voltage-clamp data on intrinsic currents and synaptic receptors present in the circuitry. The model suggests that corticothalamic feedback must operate on the thalamus mainly through excitation of GABAergic RE neurons, therefore recruiting relay cells essentially through inhibition and rebound. We provide experimental evidence for such dominant inhibition in the lateral posterior nucleus. In these conditions, the model shows that cortical discharges optimally evoked thalamic oscillations. This feature is essential to the present cellular mechanism and is also consistently observed experimentally. The model further shows that, with this type of corticothalamic feedback, cortical discharges recruited large areas of the thalamus because of the divergent cortex-to-RE and RE-to-TC axonal projections. Consequently, the thalamocortical network generated patterns of oscillations and synchrony similar to in vivo recordings. The model also emphasizes the important role of the modulation of the Ih current by calcium in TC cells. This property conferred a relative refractoriness to the entire network, a feature also observed experimentally, as we show here. Further, the same property accounted for various spatiotemporal features of oscillations, such as systematic propagation after low-intensity cortical stimulation, local oscillations, and more generally, a high variability in the patterns of spontaneous oscillations, similar to in vivo recordings. We propose that the large-scale synchrony of spindle oscillations in vivo is the result of thalamocortical interactions in which the corticothalamic feedback acts predominantly through the RE nucleus. Several predictions are suggested to test the validity of this model.

Mechanisms of long-lasting hyperpolarizations underlying slow sleep oscillations in cat corticothalamic networks

1. To explore the nature of the long-lasting hyperpolarizations that characterize slow oscillations in corticothalamic circuits in vivo, intracellular recordings were obtained under ketamine-xylazine anaesthesia from cortical (Cx) cells of the cat precruciate motor cortex, thalamic reticular (RE) cells from the rostrolateral sector, and thalamocortical (TC) cells from the ventrolateral (VL) nucleus. 2. Measurements in the three cell types showed input resistance (Rin) to be highest during the long-lasting hyperpolarizations that correspond to depth-positive waves of the cortical EEG. Rin was lowest during the early phase of high-amplitude depth-negative EEG waves and increased thereafter until the next cycle of the slow oscillation. 3. Spontaneous long-lasting hyperpolarizations were compared with those evoked by dorsal thalamic stimulation. Voltage versus current (V-I) plots showed similar membrane potential (Vm) ranges and slopes for spontaneous and evoked hyperpolarizations in both Cx and RE cells. V-I plots from TC cells had similar slopes, but Vm during evoked hyperpolarizations was displaced towards more negative values. 4. Intracellular injection of constant hyperpolarizing current in Cx cells increased the amplitude of the initial part of the depolarizing plateau of the slow oscillation, but decreased the amplitude of the last part. 5. These results suggest disfacilitation to be the dominant mechanism in the membrane of cortical and thalamic cells during the spontaneous long-lasting hyperpolarizations, which shape and synchronize slow oscillations in corticothalamic networks. In Cx and RE cells, the same mechanism underlies thalamically evoked long-lasting hyperpolarizations. By contrast, evoked responses in TC cells show a strong additional hyperpolarizing factor. We propose that GABAB processes are stronger in TC than in Cx neurones, thus rendering the thalamus an easier target for absence-type epileptic phenomena through potentiation of thalamic rebound capabilities.

Spike-wave complexes and fast components of cortically generated seizures. I. Role of neocortex and thalamus

We explored the relative contributions of cortical and thalamic neuronal networks in the generation of electrical seizures that include spike-wave (SW) and polyspike-wave (PSW) complexes. Seizures were induced by systemic or local cortical injections of bicuculline, a gamma-aminobutyric acid-A (GABAA) antagonist, in cats under barbiturate anesthesia. Field potentials and extracellular neuronal discharges were recorded through arrays of eight tungsten electrodes (0.4 or 1 mm apart) placed over the cortical suprasylvian gyrus and within the thalamus. 1) Systemic injections of bicuculline induced SW/PSW seizures in cortex, whereas spindle sequences continued to be present in the thalamus. 2) Cortical suprasylvian injection of bicuculline induced focal paroxysmal single spikes that developed into full-blown seizures throughout the suprasylvian cortex. The seizures were characterized by highly synchronized SW or PSW complexes at 2-4 Hz, interspersed with runs of fast (10-15 Hz) activity. The intracellular aspects of this complex pattern in different types of neocortical neurons are described in the following paper. Complete decortication abolished the seizure, leaving intact thalamic spindles. Injections of bicuculline in the cortex of athalamic cats resulted in similar components as those occurring with an intact thalamus. 3) Injection of bicuculline in the thalamus decreased the frequency of barbiturate spindles and increased the synchrony of spike bursts fired by thalamocortical and thalamic reticular cells but did not induce seizures. Decortication did not modify the effects of bicuculline injection in the thalamus. Our results indicate that the minimal substrate that is necessary for the production of seizures consisting of SW/PSW complexes and runs of fast activity is the neocortex.

Spindle oscillation in cats: the role of corticothalamic feedback in a thalamically generated rhythm

1. Spindles represent an oscillatory activity (7-14 Hz) of the electroencephalogram (EEG) originating in the thalamus and appearing during early stages of sleep. We investigated: (i) the phase relations between thalamic and cortical neurons during this rhythm; (ii) the patterns of spindles under different anaesthetics and their modifications at various levels of the membrane potential (Vm); and (iii) the potentiating role of the corticothalamic feedback in the genesis of spindles. Intra- and extracellular recordings were performed in cats from reticular and dorsal thalamic nuclei, as well as from various cortical areas. 2. In thalamic reticular neurons, spindles were sequences of waves at 7-14 Hz, riding on a prolonged depolarizing plateau and occurring in phase with depth-negative cortical EEG waves. In thalamocortical cells, spindles consisted of inhibitory postsynaptic potentials (IPSPs) in phase with depth-positive cortical EEG waves and occasionally leading to rebound spike bursts. In cortical cells, spindle waves were rhythmic (7-14 Hz) excitatory postsynaptic potentials (EPSPs) that sometimes gave rise to action potentials. Spindles occurred in phase among thalamic reticular, thalamocortical and neocortical neurons. 3. In thalamic reticular neurons, spindle waves and their depolarizing plateaux increased in amplitude with slight cellular hyperpolarization, but at a Vm more negative than -80 or -85 mV they decreased in amplitude. No frequency alterations were observed with these Vm changes. 4. The waxing-and-waning pattern of spontaneous spindles under barbiturate anaesthesia was distinct from the waning pattern under ketamine-xylazine anaesthesia. Under all anaesthetics, spindles had a waning pattern when elicited by cortical stimuli. The amplitude of cortical-evoked spindle waves diminished with the decrease in stimulation intensity. 5. Under urethane or ketamine-xylazine anaesthesia, spindle sequences were grouped by a cortically generated slow oscillation (< 1 Hz) and were preceded by a depth-positive EEG wave that corresponded to a prolonged hyperpolarization in all three investigated (cortical, thalamic reticular, and thalamocortical) cellular types. 6. We propose that the waxing pattern of spindle oscillation is due to a progressive entrainment of units into the oscillation until a maximum number is reached, depending on the background activity in the network. The phase relations between cortical, thalamic reticular and thalamocortical neurons are ascribed to distributed excitatory signals from thalamocortical neurons to both cortical and reticular neurons at each cycle of the oscillation. In turn, cortical neurons provide a powerful drive to potentiate the genesis of thalamic spindles.

Cortical and thalamic cellular correlates of electroencephalographic burst-suppression

This experimental study on anesthetized cats used intracellular recordings of cortical, thalamocortical and reticular thalamic neurons (n = 54), as well as multi-site extracellular recordings (n = 36), to investigate the cellular correlates of EEG burst-suppression patterns, defined as alternating wave bursts and periods of electrical silence. Burst-suppression was elicited by the administration of the same or other anesthetic agents upon the background of an already synchronized EEG activity. About 95% of cortical cells entered burst-suppression, in close time-relation with EEG activity, displaying sequences of phasic depolarizing events associated with bursts of EEG waves and an electrical silence of the neuronal membrane during flat EEG epochs. The membrane potential (Vm) hyperpolarized by approximately 10 mV prior to any EEG change and the slow rhythms reflecting deep stages of anesthesia progressively disorganized with transition to burst-suppression. During flat EEG epochs, the apparent input resistance (tested through short hyperpolarizing current pulses) decreased (range 12-60%) and neuronal responsiveness to orthodromic volleys (tested by thalamic and cortical evoked excitatory postsynaptic potentials) was dramatically reduced. It is proposed that the decreased input resistance is mainly due to an increase in K+ conductances. At variance with cortical neurons, only 60-70% of thalamic cells ceased firing before overt EEG burst-suppression and were completely silent during flat periods of EEG activity. The remaining 30-40% of thalamic cells discharged rhythmic (1-4 Hz) spike bursts during periods of EEG silence. This rhythm, within the frequency range of delta waves, is generated in thalamic cells by the interplay between two of their intrinsic currents at critical levels of Vm hyperpolarization. However, with the deepening of burst-suppression, when silent EEG periods became longer than 30 sec, thalamic cells also ceased firing. The assumption that full-blown burst-suppression is achieved through virtually complete disconnection in brain circuits implicated in the genesis of the EEG is corroborated by the revival of normal cellular and EEG activities after volleys setting into action thalamic and cortical networks.

Synchronized sleep oscillations and their paroxysmal developments

The state of resting sleep is associated with a series of oscillations generated in cortical and thalamic networks. A newly discovered rhythm groups the spindle and delta sleep oscillations within slowly recurring (< 1 Hz) sequences. Multi-site, extra- and intracellular recordings provide evidence for synchronization of various classes of cell in the neocortex and thalamus during sleep oscillations that might reach paroxysmal levels similar to epileptic states. Sleep oscillations and the underlying synchronizing processes are disrupted during transition to brain arousal.

Electrophysiological properties of intralaminar thalamocortical cells discharging rhythmic (approximately 40 HZ) spike-bursts at approximately 1000 HZ during waking and rapid eye movement sleep

Thalamocortical neurons located in the large-celled district of the cat intralaminar centrolateral nucleus were found to discharge spike-bursts with unusually high frequencies (800-1000 Hz) during spindle oscillations of the electroencephalogram. In chronically implanted animals, similar spike-bursts were also fired during wakefulness and rapid eye movement sleep, two behavioral states in which other thalamocortical neurons tonically fire single spikes. Such high-frequency spike-bursts recurred with a fast rhythm of 20-40 Hz during waking and rapid eye movement sleep. Intracellular recordings under barbiturate anesthesia showed that, during spindle oscillations, the spike-bursts of intralaminar neurons are generated by brief low-threshold spikes with a much shorter refractory phase than in other thalamocortical cells. Depolarizing pulses from the resting membrane potential triggered fast oscillations (20-80 Hz) crowned by short high-frequency (800-1000 Hz) spike-bursts. During the inter-spindle epochs, the "tonic" firing of these neurons was, in fact, a fast oscillation (30-40 Hz) of the membrane potential leading to single spikes or spike-doublets. Autocorrelograms computed from inter-spindle epochs, at relatively depolarized levels, confirmed the presence of multiple peaks at this fast rhythm. The properties of these neurons make them well suited for the distribution of fast rhythms during arousal and rapid eye movement sleep over the cerebral cortex.

Glucose-induced oscillations of intracellular Ca2+ concentration resembling bursting electrical activity in single mouse islets of Langerhans

Intracellular Ca2+ levels were monitored in single, acutely isolated mouse islets of Langerhans by dual emission Indo-1 fluorometry. High-frequency (3.1 min-1) [Ca2+]i oscillations with a brief rising time (1-2 s) and 10 s half-width ('fast' oscillations) were detected in 11 mM glucose. Raising the glucose concentration to 16.7 mM increased the duration of these oscillations, which were otherwise absent in 5.5 mM glucose. [Ca2+]i waves of lower frequency (0.5 min-1) and longer rising time ('slow' oscillations) were also recorded. The data indicate that "fast" oscillations are directly related to beta-cell bursting electrical activity, and suggest the existence of extensive networks of electrically coupled cells in the islet.

A model of spindle rhythmicity in the isolated thalamic reticular nucleus

1. The oscillatory properties of the isolated reticular (RE) thalamus were modeled with the use of compartmental models of RE cells. Hodgkin-Huxley type kinetic models of ionic channels were derived from voltage- and current-clamp data from RE cells. Interactions between interconnected RE cells were simulated with the use of a kinetic model of gamma-aminobutyric acid (GABA) inhibitory synapses. 2. The intrinsic bursting properties of RE cells in the model were due to the presence of a low-threshold Ca2+ current and two Ca(2+)-activated currents. The properties of these model RE cells were compared with RE neurons recorded intracellularly in vivo in cats. 3. Model RE cells densely interconnected with GABAA synapses produced synchronous oscillations at a frequency close to that of spindles (7-14 Hz). Networks of RE neurons organized in a two-dimensional array with only proximal connectivity also exhibited synchronized oscillations in the spindle range. In addition, the proximally connected network showed periods of high and low synchronicity, giving rise to waxing and waning oscillations in the population of RE cells. 4. The spatiotemporal behavior of the network was investigated during waxing and waning oscillations. The waxing and waning emerged as an alternation between periods of desynchronized and synchronized activity, corresponding to periods of irregular and coherent spatial activity. During synchronized periods, the network displayed propagating coherent waves of synchronous activity that had a tendency to form spirals. 5. Networks of model RE neurons fully connected through GABAB synapses exhibited perfectly synchronous oscillations at lower frequencies (0.5-1 Hz), but two-dimensional networks with proximal GABAB connectivity failed to synchronize. 6. These simulations demonstrate that networks of model neurons that include the main intrinsic currents found in RE cells can generate waxing and waning oscillatory activity similar to the spindle rhythmicity observed in the isolated RE nucleus in vivo. The model reveals the interplay between the intrinsic rhythmic properties of RE cells and the fast synaptic interactions in organizing synchronized rhythmicity.

N-methyl-d-aspartic acid receptor antagonist-induced frequency oscillations in mice recreate pattern of electrophysiological deficits in schizophrenia

INTRODUCTION

Electrophysiological responses to auditory stimuli have provided a useful means of elucidating mechanisms and evaluating treatments in psychiatric disorders. Deficits in gating during paired-click tasks and lack of mismatch negativity following deviant stimuli have been well characterized in patients with schizophrenia. Recently, analyses of basal, induced, and evoked frequency oscillations have gained support as additional measures of cognitive processing in patients and animal models. The purpose of this study is to examine frequency oscillations in mice across the theta (4-7.5 Hz) and gamma (31-61 Hz) bands in the context of N-methyl-d-aspartic acid receptor (NMDAR) hypofunction and dopaminergic hyperactivity, both of which are thought to serve as pharmacological models of schizophrenia.

EXPERIMENTAL PROCEDURES

Electroencephalograms (EEG) were recorded from mice in five treatment groups that consisted of haloperidol, risperidone, amphetamine, ketamine, or ketamine plus haloperidol during an auditory task. Basal, induced and evoked powers in both frequencies were calculated.

RESULTS

Ketamine increased basal power in the gamma band and decreased the evoked power in the theta band. The increase in basal gamma was not blocked by treatment with a conventional antipsychotic. No other treatment group was able to fully reproduce this pattern in the mice.

CONCLUSIONS

Ketamine-induced alterations in EEG power spectra are consistent with abnormalities in the theta and gamma frequency ranges reported in patients with schizophrenia. Our findings support the hypothesis that NMDAR hypofunction contributes to the deficits in schizophrenia and that the dopaminergic pathways alone may not account for these changes.

Synaptic responsiveness of cortical and thalamic neurones during various phases of slow sleep oscillation in cat

1. The fluctuations during various phases of the slow sleep oscillation (< 1 Hz) in synaptic responsiveness of motor cortical (Cx), thalamic reticular (RE) and thalamocortical (TC) neurones were investigated intracellularly in cats under ketamine-xylazine anaesthesia. Orthodromic responses to stimuli applied to brachium conjunctivum (BC) axons and corticothalamic pathways were studied. The phases of slow oscillation consist of a long-hyperpolarized, followed by a sharp depth-negative EEG deflection and a series of faster waves that are associated with the depolarization of Cx and RE neurones, while TC cells display a sequence of IPSPs within the spindle frequency. 2. BC-evoked bisynaptic excitatory postsynaptic potentials (EPSPs) in Cx and RE neurones were drastically reduced in amplitude during the long-lasting hyperpolarization and the early part of the depolarizing phase. By contrast, the BC-evoked monosynaptic EPSPs of TC cells were not diminished during the depth-positive EEG wave, but the hyperpolarization during this phase of the slow oscillation prevented TC neurones transferring prethalamic signals to the cortex. 3. At variance with the diminished bisynaptic EPSPs evoked in response to BC stimuli during the long-lasting hyperpolarization, Cx-evoked monosynaptic EPSPs in Cx cells increased linearly with hyperpolarization during this phase of the slow oscillation. Similarly, the amplitudes of Cx-evoked EPSPs in RE and TC cells were not diminished during the long-lasting hyperpolarization. 4. The diminished responsiveness of Cx and RE neurones to prethalamic volleys during the long-lasting hyperpolarization is attributed to gating processes at the level of TC cells that, because of their hyperpolarization, do not transfer prethalamic information to further relays.

Electrophysiological properties of cat reticular thalamic neurones in vivo

1. The electrophysiological properties of neurones of the reticular thalamic (RE) nucleus were studied in acutely prepared cats under urethane anaesthesia. 2. Two main types of neuronal firing were recorded. At the resting membrane potential (-60 to -65 mV) tonic repetitive firing was elicited when the cell was activated synaptically or by current injection. From membrane potentials more negative than -75 mV, synaptic or direct stimulation generated a burst of action potentials. 3. The burst of RE cells consisted of a discharge of four to eight spikes riding on a slowly growing and decaying depolarization. The discharge rate during the burst showed a characteristic increase, followed by a decrease in frequency. 4. The burst response behaved as a graded phenomenon, as its magnitude was modulated by changing the intensity of the synaptic volley or the intensity of the injected current. 5. Spike-like small potentials presumably of dendritic origin occurred spontaneously and were triggered by synaptic or direct stimulation. They were all-or-none, voltage-dependent events. We postulate that these spikes originate in several hot spots in the dendritic arbor, with no reciprocal refractoriness and may generate multi-component depolarizations at the somatic level. 6. Excitatory postsynaptic potentials (EPSPs) evoked by internal capsule stimulation consisted of two components, the late one being blocked by hyperpolarization. Such compound EPSPs were followed by a period of decreased excitability during which a second response was diminished in amplitude. 7. A series of depolarizing waves at the frequency range of spindle oscillations was triggered by internal capsule stimulation. The individual depolarizing waves constituting the spindle oscillation gradually decreased in amplitude when decreasing the intensity of the stimulation. 8. These results, showing that RE cells are endowed with an excitable dendritic tree and a graded bursting behaviour, support the proposed role of RE nucleus as the generator and synchronizer of spindle rhythmicity.

Intracellular and computational characterization of the intracortical inhibitory control of synchronized thalamic inputs in vivo

We investigated the presence and role of local inhibitory cortical control over synchronized thalamic inputs during spindle oscillations (7-14 Hz) by combining intracellular recordings of pyramidal cells in barbiturate-anesthetized cats and computational models. The recordings showed that 1) similar excitatory postsynaptic potential (EPSP)/inhibitory postsynaptic potential (IPSP) sequences occurred either during spindles or following thalamic stimulation; 2) reversed IPSPs with chloride-filled pipettes transformed spindle-related EPSP/IPSP sequences into robust bursts with spike inactivation, resembling paroxysmal depolarizing shifts during seizures; and 3) dual simultaneous impalements showed that inhibition associated with synchronized thalamic inputs is local. Computational models were based on reconstructed pyramidal cells constrained by recordings from the same cells. These models showed that the transformation of EPSP/IPSP sequences into fully developed spike bursts critically needs a relatively high density of inhibitory currents in the soma and proximal dendrites. In addition, models predict significant Ca2+ transients in dendrites due to synchronized thalamic inputs. We conclude that synchronized thalamic inputs are subject to strong inhibitory control within the cortex and propose that 1) local impairment of inhibition contributes to the transformation of spindles into spike-wave-type discharges, and 2) spindle-related inputs trigger Ca2+ events in cortical dendrites that may subserve plasticity phenomena during sleep.

Cortically-induced coherence of a thalamic-generated oscillation

Oscillatory patterns in neocortical electrical activity show various degrees of large-scale synchrony depending on experimental conditions, but the exact mechanisms underlying these variations of coherence are not known. Analysis of multisite local field potentials revealed that the coherence of spindle oscillations varied during different states. During natural sleep, the coherence was remarkably high over cortical distances of several millimeters, but could be disrupted by artificial cortical depression, similar to the effect of barbiturates. Possible mechanisms for these variations of coherence were investigated by computational models of interacting cortical and thalamic neurons, including their intrinsic firing patterns and various synaptic receptors present in the circuitry. The model indicates that modulation of the excitability of the cortex can affect spatiotemporal coherence with no change in the thalamus. The highest level of coherence was obtained by enhancing the excitability of cortical pyramidal cells, simulating the action of neuromodulators such as acetylcholine and noradrenaline. The underlying mechanism was due to cortex-thalamus-cortex loops in which a more excitable cortical network generated a more powerful and coherent feedback onto the thalamus, resulting in highly coherent oscillations, similar to the properties measured during natural sleep. In conclusion, these experiments and models are compatible with a powerful role for the cortex in triggering and synchronizing oscillations generated in the thalamus, through corticothalamic feedback projections. The model suggests that intracortical mechanisms may be responsible for synchronizing oscillations over cortical distances of several millimeters through cortex-thalamus-cortex loops, thus providing a possible cellular mechanism to explain the genesis of large-scale coherent oscillations in the thalamocortical system.

Intracellular evidence for incompatibility between spindle and delta oscillations in thalamocortical neurons of cat

Recent studies have revealed that the thalamus does not only generate spindle oscillations (7-14 Hz), but that it also participates in the genesis of a slower (less than 4 Hz) rhythm within the frequency range of delta waves on the electroencephalogram. In thalamic cells, delta is an intrinsic oscillation consisting of low-threshold spikes alternating with afterhyperpolarizing potentials. It is known from electroencephalographic recordings in humans and animals that slow or delta waves prevail during late sleep stages, whereas spindle oscillations are characteristic for the early stages of sleep. We studied the dependence of spindles and delta oscillations on membrane potential, as well as the effects of spindles on delta oscillations, in thalamocortical neurons of cats under urethane anesthesia and in cerveau isolé preparations (low collicular transections). Spindles appeared at membrane potentials between -55 and -65 mV, whereas delta oscillations occurred by bringing the membrane potential between -68 and -90 mV. Spindles either evoked by cortical stimulation or occurring spontaneously in cerveau isolé preparations prevented delta oscillations. This effect was probably due to the increase in membrane conductance associated with spindles. Barbiturates also blocked delta activity in thalamocortical neurons, probably through the same mechanism. A certain degree of incompatibility between spindles and delta rhythms in thalamocortical cells may explain the prevalence of these two types of oscillations during different stages of sleep with synchronization of the electroencephalogram.

Synchronization of low-frequency rhythms in corticothalamic networks

We have investigated the degree of synchronization between cortical, thalamic reticular and thalamocortical neurons of cats during low-frequency (< 15 Hz) sleep-like oscillations, as they appear under anaesthesia. We have also studied the effects exerted by cortical stimulation on the synchronization among thalamic units. Parallel experiments [Steriade et al. (1996) J. Neurosci. 16, 392-417] in this laboratory have demonstrated the similarity between the slow oscillation (< 1 Hz) under ketamine-xylazine anaesthesia and that occurring during the natural state of resting sleep. Spontaneous activity was recorded simultaneously, with independent microelectrodes, from groups of two to five physiologically identified neurons. The rhythmicity of spontaneous activity and the temporal relations between cellular discharges were statistically evaluated by auto- and crosscorrelation techniques. We have found no topography in the distribution of synchronization between thalamic reticular and thalamocortical cells. Only the slow, cortical-generated oscillation (< 1 Hz) displayed a stable frequency and correlation among groups of cortical and thalamic cells. The other two sleep oscillations (thalamic-generated spindles at 7-14 Hz and clock-like delta at 1-4 Hz) fluctuated in frequency and the degree of correlation between neurons varied. Cortical volleys entrained and synchronized thalamic cells, and triggered synchronized spindling in the thalamus. These results extend for large populations of cortical and thalamic neurons the phase relations found in intracellular recordings.

The relationship between glucose-induced K+ATP channel closure and the rise in [Ca2+]i in single mouse pancreatic beta-cells

1. Intracellular calcium [Ca2+]i and channel activity were simultaneously recorded in single, dissociated mouse beta-cells kept in culture for 1-3 days. [Ca2+]i was estimated from microfluorometric ratio methods using Indo-1. Channel activity was measured using the cell-attached configuration of the patch-clamp technique. 2. At low glucose concentrations (0.3 mM), resting K+ATP channel activity was prevalent. Increasing glucose up to 16 mM, produced a gradual decrease in K+ATP channel activity over a time course of 90-120 s (temperature = 23 degrees C) and an increase in [Ca2+]i. 3. In the majority of experiments, glucose elicited biphasic action currents (action potentials) which preceded the rise in [Ca2+]i. There was a close correlation between spike frequency and the levels of [Ca2+]i. 4. The sulphonylurea tolbutamide (1 mM) blocked K+ATP channels in 10-20 s. K+ATP channel blockade was associated with a quick rise in [Ca2+]i. 5. When K+ATP channel activity was stimulated in the presence of diazoxide (100 microM), increasing the glucose concentration from 3 to 16 mM produced a decrease in [Ca2+]i. Only when diazoxide was removed did glucose produce an increase in [Ca2+]i. 6. In a small population of cells, glucose (16 mM) produced a small decrease in K+ATP channel activity but not an increase in [Ca2+]i. In such cells, tolbutamide blocked K+ATP channels and produced an increase in [Ca2+]i. 7. These results demonstrate a close correlation between K+ATP channel activity and [Ca2+]i in beta-cells. The findings are consistent with the model in which glucose metabolism produces a rise in [Ca2+]i through the blockade of K+ATP channels, membrane depolarization and calcium current activation.

Bursting and tonic discharges in two classes of reticular thalamic neurons

1. Two types of cat reticular (RE) thalamic cells were disclosed by means of intracellular recordings under urethan anesthesia. The RE neurons were identified by their typical depolarizing spindle oscillations in response to synchronous stimulation of the internal capsule. 2. In type I neurons (n = 41), depolarizing current pulses induced tonic firing at the resting or slightly depolarized membrane potential (Vm) and triggered high-frequency spike bursts at a Vm more negative than -75 mV. As well, these cells discharged rebound bursts at the break of a hyperpolarizing current pulse. Internal capsule stimulation elicited spindle sequences made off by depolarizing waves giving rise to spike bursts. 3. Type II cells (n = 9) did not discharge spike bursts to large depolarizing current pulses even when the Vm reached -100 mV, nor did they fire rebound bursts after long-lasting hyperpolarizing current pulses or spike bursts riding on the rhythmic depolarizing components of spindle sequences. 4. Compared with type I cells, type II cells showed less frequency accommodation during tonic firing. The latter neuronal class discharged at high frequencies (40 Hz) with slight DC depolarization, approximately 8-10 Hz at the resting Vm, and no underlying synaptic or subthreshold oscillatory events could be detected when the firing was blocked by DC hyperpolarization. 5. The presence of two cell classes in the RE nucleus challenges the common view that this nucleus consists of a single neuronal class. We suggest that a different set of conductances is present in type II RE neurons, thus preventing the low-threshold Ca2+ current from dominating the behavior of these cells.

Modeling the control of reticular thalamic oscillations by neuromodulators

Compartmental models of thalamic reticular (RE) neurons were investigated based on current-clamp and voltage-clamp data. Spontaneous oscillations in the model arise from the interaction between inhibitory synaptic currents and the rebound burst of RE cells. These oscillations critically depend on the level of the resting membrane potential. A network of RE neurons can be switched between silent and sustained oscillatory behavior by modulating a leak potassium current through neuromodulatory synapses. These results suggest that neuromodulators, such as noradrenaline, serotonin and glutamate, can exert a decisive control over the oscillatory activity of systems of RE cells. The model may explain why the isolated RE nucleus oscillates spontaneously in vivo but not in vitro.

Dynamic interactions determine partial thalamic quiescence in a computer network model of spike-and-wave seizures

In vivo intracellular recording from cat thalamus and cortex was performed during spontaneous spike-wave seizures characterized by synchronously firing cortical neurons correlated with the electroencephalogram. During these seizures, thalamic reticular (RE) neurons discharged with long spike bursts riding on a depolarization, whereas thalamocortical (TC) neurons were either entrained into the seizures (40%) or were quiescent (60%). During quiescence, TC neurons showed phasic inhibitory postsynaptic potentials (IPSPs) that coincided with paroxysmal depolarizing shifts in the simultaneously recorded cortical neuron. Computer simulations of a reciprocally connected TC-RE pair showed two major modes of TC-RE interaction. In one mode, a mutual oscillation involved direct TC neuron excitation of the RE neuron leading to a burst that fed back an IPSP into the TC neuron, producing a low-threshold spike. In the other, quiescent mode, the TC neuron was subject to stronger coalescing IPSPs. Simulated cortical stimulation could trigger a transition between the two modes. This transition could go in either direction and was dependent on the precise timing of the input. The transition did not always follow the stimulation immediately. A larger, multicolumnar simulation was set up to assess the role of the TC-RE pair in the context of extensive divergence and convergence. The amount of TC neuron spiking generally correlated with the strength of total inhibitory input, but large variations in the amount of spiking could be seen. Evidence for mutual oscillation could be demonstrated by comparing TC neuron firing with that in reciprocally connected RE neurons. An additional mechanism for TC neuron quiescence was assessed with the use of a cooperative model of gamma-aminobutyric acid-B (GABA(B))-mediated responses. With this model, RE neurons receiving repeated strong excitatory input produced TC neuron quiescence due to burst-duration-associated augmentation of GABA(B) current. We predict the existence of spatial inhomogeneity in apparently generalized spike-wave seizures, involving a center-surround pattern. In the center, intense cortical and RE neuron activity would be associated with TC neuron quiescence. In the surround, less intense hyperpolarization of TC neurons would allow low-threshold spikes to occur. This surround, an "epileptic penumbra," would be the forefront of the expanding epileptic wave during the process of initial seizure generalization. Therapeutically, we would then predict that agents that reduce TC neuron activity would have a greater effect on seizure onset than on ongoing spike-wave seizures or other thalamic oscillations.

Block of transmitter release by botulinum C1 action on syntaxin at the squid giant synapse

Electrophysiological, morphological, and biochemical approaches were combined to study the effect of the presynaptic injection of the light chain of botulinum toxin C1 into the squid giant synapse. Presynaptic injection was accompanied by synaptic block that occurred progressively as the toxin filled the presynaptic terminal. Neither the presynaptic action potential nor the Ca2+ currents in the presynaptic terminal were affected by the toxin. Biochemical analysis of syntaxin moiety in squid indicates that the light chain of botulinum toxin C1 lyses syntaxin in vitro, suggesting that this was the mechanism responsible for synaptic block. Ultrastructure of the injected synapses demonstrates an enormous increase in the number of presynaptic vesicles, suggesting that the release rather than the docking of vesicles is affected by biochemical lysing of the syntaxin molecule.

State-dependent fluctuations of low-frequency rhythms in corticothalamic networks

We have studied the variations in the degree of correlated firing within the low-frequency sleep rhythms (< 15 Hz) between cortical, thalamic reticular and thalamocortical neurons during changes in the amplitude and frequency of brain electrical activity in anaesthetized cats. Extracellular discharges of neuronal groups of two to five physiologically identified cortical and thalamic units were recorded simultaneously with independent microelectrodes. The firing patterns and the temporal correlation between spike-trains were evaluated by auto- and crosscorrelograms. Although the animals were under deep anaesthesia, additional doses of the same or different anaesthetics were able to alter the electroencephalographic pattern, inducing waves with higher amplitude. Similar transitions occurred spontaneously. We found that the presence of rhythmic behaviour in cells of corticothalamic networks, as well as their degree of correlated firing, was extremely sensitive to even slight alterations in the state of the electroencephalogram. Cells belonging to the same functional system, but located distantly, became highly synchronized upon the increased amplitude of brain waves. Thus, an electroencephalogram characterized by slow waves corresponds to a state of rhythmic and correlated firing among cortical and thalamic neurons. The highly coherent activity during sleep patterns transcends the borders which limit the functioning during the waking brain.