*Department of Anaesthesiology, University of Tübingen, D-72072 Tübingen and Max Planck Institute for Biological Cybernetics, D-72076 Tübingen, Germany*Address for correspondence
Abstract
Br J Anaesth 2002; 89: 10211
Keywords: brain, cerebral cortex; brain, GABAA receptor; monitoring, electroencephalography; sleep
Brain slices are well-established tools in neuroscience research. In the last decade scientists succeeded in isolating viable brain slices from many different regions of the mammalian central nervous system. The brain slice preparation was introduced by Henry MacIlwain in the 1950s.40 45 46 Milestones in the development of this technique are listed in Table 1. Brain slices can be viewed as intact local microcircuits, lacking synaptic inputs from other parts of the central nervous system. In using this type of preparation, drug actions on specific types of neurones have been investigated on the network, cellular, and molecular level. Studies on brain slices provided interesting insights into the mechanisms by which general anaesthetics affect cortical neurones. In the last decade, brain slices also proved to be helpful for analysing patterns of neuronal activity, observed during the anaesthetic state in experimental animals. There is increasing interest in uncovering the contribution of specific local microcircuits to the overall changes in neuronal activity evoked by general anaesthetics. Electroencephalographic (EEG) recordings have shown that anaesthetic agents dramatically alter the firing mode of cortical neurones, even when applied in a range of rather low concentrations.20 68 As well as compound-specific actions, almost each anaesthetic that has been investigated transfers high-frequency low-voltage EEG activity, present during the awake state and rapid eye movement (REM) sleep, into low-frequency high-voltage activity. The latter is commonly taken as evidence for the presence of a hypnotic, delta sleep-like state. However, interpretation of EEG recordings turned out to be rather difficult, as they provide a complex correlate of the summed synaptic activity of neurones located in the upper layers of the cerebral cortex.53 As yet, the physiological mechanisms that underlie EEG activity during anaesthesia remain largely unknown. Recent imaging studies on human subjects demonstrated that, besides altering synchronized activity, some general anaesthetics strongly depress cortical metabolism and blood flow.1 2 4 5 24 Again, the physiological processes underlying these phenomena remain to be elucidated. In this article, anaesthetic actions in brain slices are discussed. Experimental findings will be related to recent work on living animals and human subjects.
|
What happens in the cerebral cortex when anaesthesia is commenced and patients lose consciousness? A considerable number of studies, making use of different methodological approaches, have tackled this question. For example, it has been shown that in lightly anaesthetized subjects the metabolism of most cortical areas is decreased by roughly 50%.1 2 4 5 24 Furthermore, the transition from wakefulness to unresponsiveness is accompanied by a depression of high-frequency electrical activity in the gamma band.69 Low-frequency high-voltage activity becomes more pronounced as anaesthesia is deepened.20 As with many anaesthetics, unresponsiveness occurs at drug concentrations far below those abolishing painful stimuli-induced movements, some aspects of information processing in the cerebral cortex seem to be rather sensitive to anaesthetic treatment. How can this be explained? Do cortical cells themselves possess molecular targets, mediating neuronal depression and hypnosis? A number of arguments indeed support this view. There is now broad agreement that GABAA and NMDA receptors are important sites of general anaesthetic actions.7 16 25 33 67 The density of these receptors in the cerebral cortex is higher than in most other parts of the central nervous system. Furthermore, a role of cortical GABAA receptors in hypnosis is also suggested by the action of sedative drugs. Benzodiazepines are well known to cause their therapeutic effects by enhancing GABAA receptor-function in a rather specific manner. Recently, evidence has been provided that benzodiazepine-induced sedation in mice is mediated via a GABAA receptor-subtype largely restricted to the cerebral cortex.47 57 Taken together, these findings indicate that hypnosis is to a substantial degree caused by drug actions on neurones located in the cortex.
However, in order to explain the hypnotic aspect of the anaesthetic state, many textbooks and review articles refer to a different theory, emphasizing similarities between anaesthesia and natural sleep. Pronounced high-voltage low-frequency EEG activity is present during the non-REM stages of natural sleep. The pioneering experiments of Moruzzi and Magoun demonstrated that the states of wakefulness and sleep as well as their electrical correlates are influenced by brain stem nuclei.48 These sub-cortical structures also might be potential sites for general anaesthetic actions.10 21 37 50 55 It may be that general anaesthetics create a brain state similar to delta sleep by modulating arousal systems.3 19 49
In summary, there are good reasons for assuming that anaesthetic-induced synchronization and depression of neuronal activity in the cerebral cortex involve sub-cortical arousal systems. However, as discussed above, we are also faced with data strongly supporting the possibility that direct actions on cortical neurones come into play. How do we evaluate the importance of different potential pathways of drug action? How do we establish a more precise hypothesis that distinguishes between cortical and sub-cortical mechanisms? What type of experiments can be conducted to elucidate the specific contributions made by different local networks? A major problem arises from the fact that distributed local microcircuits are extensively interconnected in intact brains: if we assume that a particular part of the brain, say the neocortex (termed network B in the following) receives excitatory input from a sub-cortical structure (network A), neuronal activity in B may be reduced either because of direct effects on drug targets located in B, or because of a decrease in excitatory drive provided by A. Thus, it is necessary to study drug effects in isolated local networks, which do not receive synaptic input from different brain areas. If it turns out that network B is equally sensitive to drug treatment, regardless of whether synaptic input provided by A is present or not, B is most probably the substrate of drug action. If B is rendered insensitive by removing synaptic input provided by A, the more important effects take place in network A.
At this point, the topics addressed above have been considered in order to establish some type of conceptual framework. Such a framework seems to be helpful for recognizing the benefit of brain slice studies and for integrating the results into a general understanding of how anaesthetics work. The research discussed in the following is centred on work dealing with the question of how general anaesthetics alter activity patterns in hippocampal and neocortical brain slices. The focus is on drug actions occurring at the network level.
Electrical correlates of neuronal network activity in vitro and in vivo
It is commonly assumed that anaesthetic-induced unconsciousness is causally related to changes in the firing characteristics of cortical neurones. Thus, monitoring the activity patterns of these cells takes a central place in understanding the physiological processes related to the hypnotic state. Today, different methods are in use. Some basic information concerning measurements of electrical brain activity is summarized in Figure 1. In the following I will consider what particular aspect of neuronal activity can be obtained by the specific methodological approaches.
|
|
Band-pass filtering the recordings at 0.1100 Hz removes action potentials: the residual voltage changes display correlated synaptic activity. This type of signal is commonly termed the local field potential (LFP) or the micro-EEG. However, in contrast to the EEG recorded from humans and animals, the LFP integrates the synaptic activity over much fewer neurones: thus the spatial resolution is far better.
How are the LFP and action potential firing related in the time domain? If oscillatory synaptic activity, for example in the gamma or delta range is present in the LFP, it is most commonly also evident in multi-unit firing. If action potential firing and LFPs are recorded simultaneously, a stable-phase relationship can be observed in most cases. However, LFP recordings do not provide information about changes in the average firing rates of cortical neurones.
In the past few years, imaging studies have shown that at light levels of anaesthesia, just sufficient to render human subjects unresponsive, cortical metabolism and blood flow are depressed by as much as 3050% compared with the awake state. Quantitatively, these findings fit well with recent recordings of the mean firing rates in rats.26 Taken together, it seems that the depression of cortical metabolism induced by anaesthetic agents is related to a strong decrease in the mean firing rates and synaptic activity. Because of the poor time resolution, imaging studies neither provide information about the dominant rhythms nor about the degree of synchronization characterizing neuronal activity during the awake or anaesthetized state.
Synchronized cortical activity in the theta and delta range
Most anaesthetic agents alter the EEG activity patterns in a similar, although not identical way. The study of MacIver and collaborators is a good example for illustrating commonly observed hallmarks of anaesthetic effects, occurring in a range of clinically relevant concentrations.41 42 The key results are summarized in Table 3. At different thiopental concentrations, the authors correlated the behavioural state in the rat with the corresponding EEG pattern and the molecular mechanisms of anaesthetic action. They observed increased power in all frequency bands of the EEG (a pattern termed activation) during sedation, delta oscillations during hypnosis, and a pattern termed burst suppression during anaesthesia. Burst suppression is characterized by episodes of low-frequency high-voltage activity, which are separated by periods of neuronal silence. Very similar concentration-dependent effects appear with a number of barbiturates and volatile anaesthetics in animals and humans. The finding that general anaesthetics induce delta activity in the EEG has prompted the suggestion that the brain states of anaesthesia and natural sleep are comparable with regard to several aspects. This hypothesis is mainly based on the fact that synchronized cortical delta activity is present during non-REM sleep. A number of studies on sleep mechanisms provided evidence that synchronized cortical delta activity originates in the thalamus. During delta sleep, cortical neurones seem to be driven and synchronized by thalamic relay neurones.44 In this sleep state, thalamic relay cells fire rhythmic bursts of fast sodium action potentials, as indicated in Figure 2. Ca2+ currents play a major role in generating the rhythm.64 In the awake state and during REM sleep, Ca2+ channels remain inactivated, as relay neurones receive a strong depolarizing synaptic input originating in brain stem nuclei. Some of the latter are involved in arousal and in controlling the sleepwake cycle. With a view on the mechanism thought to underlie natural sleep, cortical delta rhythms during anaesthesia could be explained by a depression of neuronal activity at the sub-cortical level. Assuming that general anaesthetics reduce the excitability of thalamic relay cells either by direct actions on these neurones or by reducing their synaptic input, it seems possible that the cells are forced to enter the burst mode, thereby producing synchronized delta activity in the thalamo-cortical loop. In this scenario, anaesthetic-induced delta activity is largely related to direct or indirect drug actions on thalamic relay neurones.
|
|
In rat neocortical slices a burst suppression-like pattern can be observed at thiopental concentrations somewhat higher than those shifting theta oscillations towards the delta range. This finding indicates that burst suppression involves a cortical site of action. In fact, recordings in the neocortex in experimental animals, before and after removing the ascending inputs by cutting along the white matter, demonstrated that burst suppression survives this procedure.34 66
Anaesthetic concentrations exceeding those causing burst suppression fully depress neuronal activity in the cortex: in vivo, an isoelectric EEG is seen during this state. In vitro, neither spontaneous changes in the LFP nor action potential activity can be observed.
What are the molecular mechanisms underlying the different stages of anaesthesia described above? Do different patterns of network activity result from anaesthetic modulation at a single molecular locus? Or, alternatively, does a multi-site theory of anaesthetic action provide a better explanation of the experimental data mentioned so far? MacIver and co-workers report that at 10 µmol litre1 thiopental the period of spontaneous theta oscillations is increased by about 2- to 3-fold.41 42 The same concentration alters the kinetic properties of GABAA receptor-mediated synaptic events, thus causing a 3-fold increase in inhibitory postsynaptic current (IPSC) decay time, leaving the amplitude of spontaneous IPSCs unchanged. At anaesthetic concentrations causing burst suppression, besides the effects on IPSC decay time, neurones were tonically inhibited. This inhibition most probably resulted from direct activation of GABAA receptors, which was consistent with the finding that the GABAA receptor-agonist muscimol induced a similar pattern of network activity. The suggestion that lengthening of IPSC decays vs tonic inhibition produce different effects on the network level has been supported by recent investigations. In cultured neocortical slices, rhythmic network activity was induced by removing the Mg2+ block of NMDA receptors. It was observed that benzodiazepines, which exclusively lengthened IPSC decay time, reduced neuronal activity during network bursting without altering the silent periods between bursts.6 In contrast, volatile anaesthetics and GABAA receptor-agonists prolonged intra-burst periods.8
What happens at even higher concentrations? At 70 µmol litre1, thiopental network activity is completely depressed.41 42 At this concentration the anaesthetic affected, in addition to GABAA receptors, glutamatergic synaptic transmission. In summary, the data suggest that different patterns of network activity evoked by anaesthetic treatment are indicative of distinct concentration-dependent actions on the molecular level, consistent with a multi-site theory of anaesthetic action.
Gamma oscillations
Cognitive tasks such as recognizing a persons face, distinguishing between a dog and a pig, or answering a question involve the activity of many neurones, widely distributed over the cortex. Obviously, there is a high degree of specialization with regard to the type of information processed by neurones in different parts of the brain. For example, in the monkeys visual system more than 30 areas have been shown to analyse specific features of sensory input provided by the retina. At this point the binding problem might come into play.56 How does the brain know that aspects of colour, shape and movement, represented in different parts of the neocortex, belong to the same object? To solve the binding problem, Singer and colleagues proposed a temporal code.23 60 62 Their hypothesis states that features of one object, represented by different neurones in different neocortical areas, are linked together by synchronous neuronal activity. The experimental observation that cognitive functions are in many instances accompanied by high-frequency oscillatory activity in the gamma range brought up the idea that features belonging to the same object are coded by neurones firing at the same phase of the rhythm. It was speculated that conscious awareness is intimately related to this type of neuronal coding.52 Although the temporal coding hypothesis seems to provide an elegant solution for the binding problem, it is far from being accepted by the scientific community. The issue is discussed in a rather controversial way.30 31 59 61 Many researchers doubt that the temporal correlation hypothesis is valid. Others have remarked that oscillatory activity in the gamma range might be a highly unspecific feature of cortical networks, an epi-phenomenon of information processing that does not code anything.
The temporal coding hypothesis also attracted researchers in the field of anaesthesia. Does it offer a simple explanation of how general anaesthetics cause unconsciousness? The central argument might read as follows. If anaesthetics disturb or depress gamma oscillatory activity in the brain, this should lead to a breakdown of object representations and consciousness.43 52 58 Although this idea is attractive, it is, because of the lack of unambiguous data, still speculative. However, investigating anaesthetic effects on gamma oscillations seems to be promising, even if the temporal coding hypothesis turns out to be wrong. In analysing anaesthetic actions on brain rhythms, a great deal of interesting data concerning their molecular and network mechanisms can be gained. Furthermore, brain rhythms are observed in humans, experimental animals, and brain slices. Assuming that the molecular and network mechanisms producing a specific rhythm are similar, comparative studies may help to close the gap between anaesthetic effects reported in living organisms and brain slices.
How does the cortex generate gamma oscillations? It has been hypothesized that gamma oscillations result from intrinsic membrane properties of cortical cells. Gray and colleagues, for example, have investigated so-called chattering cells located in the upper layers of the visual neocortex of the cat.11 32 They showed that these cells develop oscillatory burst activity in response to excitatory input. They speculated that, because of the extensive horizontal synaptic connections made by chattering cells, they synchronize each other and drive many other neurones. Other investigators considered the possibility that gamma oscillations result from network interactions in the neocortex and hippocampus.36 38 39 Figure 3 illustrates further mechanisms, discussed in this context. Two models, the recurrent inhibition and mutual inhibition model share a number of properties.36 They have in common that the decay time of GABAA receptor-mediated synaptic events is a critical variable, determining the period of the rhythm. The simple relationship between the decay time of inhibitory synaptic events and the frequency of the network rhythm is the most interesting feature of these models, predicting that a 2- to 3-fold increase in the IPSC decay time should decrease the oscillation frequency of the network by half. This relationship was derived by computer modelling and confirmed experimentally in hippocampal and neocortical brain slices. In these experiments anaesthetic agents and benzodiazepines have been used as pharmacological tools for lengthening IPSC decay times in order to test model predictions.9 7274
|
Gamma and theta rhythms: what is similar, what is different?
Anaesthetic-induced changes in oscillatory network activity in the neocortex and hippocampus seem to be causally related to altered GABAA receptor-function. In the hippocampus kinetically distinct GABAA receptors co-exist.13 14 51 Fast GABAA receptors, exhibiting a half-maximal IPSC decay time of about 10 ms are located close to the somata of pyramidal cells. They were termed GABAA-fast. In contrast, slow GABAA receptors with a decay time of 60200 ms were observed on the dendrites of pyramidal cells (GABAA-slow). Interestingly, these different types of postsynaptic receptors seem to be activated by two different pools of inhibitory interneurones. Within these pools neurones synchronize their activity via chemical and electrical synaptic connections.27 GABAergic interneurones located in the stratum lacunosum-moleculare predominantly synapse on the dendrites of pyramidal cells thus activating slow GABAA receptors. In contrast, interneurones close to the stratum pyramidale predominantly make synaptic contacts close to the soma, thus opening fast GABAA receptors. The distinct pools of GABAergic interneurones interact thereby producing nested rhythmic activity at theta and gamma frequencies.15 71
Following these exciting observations, it is possible to understand how general anaesthetics may affect different types of network oscillations by acting on GABAA receptors.
It is very likely that the different kinetic properties of dendritic and somatic GABAA receptors in the hippocampus involve differences in the subunit composition. At present it is not known what subunits are related to GABAA-fast and GABAA-slow receptors. It is also unclear whether these GABAA receptor-subtypes differ in their sensitivity to general anaesthetics. However, if GABAA-slow receptors participate in producing hippocampal theta oscillations whereas GABAA-fast receptors are involved in gamma oscillations, general anaesthetics may slow both gamma and theta oscillations provided that both receptor subtypes are modulated by general anaesthetics at clinically relevant concentrations.
Anaesthetic actions on cortical metabolism and average discharge rates
General anaesthetics alter not only synchronized firing of cortical neurones. In a few studies, the effects of halothane, isoflurane, and propofol on brain metabolism and blood flow have been characterized.1 2 4 5 24 These investigations demonstrated that at drug concentrations rendering human subjects unconscious (they no longer respond to verbal commands), cerebral blood flow and metabolism is reduced by approximately 50%. When these findings were published, it was not known how to relate measurements of cerebral blood flow and metabolism to changes in spontaneous spiking and synaptic activity of neurones. Very recently, spontaneous firing of neurones in the somatosensory cortex of the rat in awake and lightly anaesthetized animals was compared. At concentrations rendering human subjects unresponsive, halothane, isoflurane and enflurane decrease spontaneous firing by 50% (H. Hentschke, personal communication, 2001). These observations fit well with the results of Gaese and Ostwald, who monitored the activity of neurones in the rat auditory cortex.26 At light levels of anaesthesia, spontaneous neuronal firing was reduced by 70% compared with the awake state.
This brings me back to whether direct anaesthetic actions on cortical neurones account for the decrease in spontaneous neuronal activity. Are neurones in cortical brain slices similarly sensitive as in vivo? To find an answer, we investigated the effects of general anaesthetics on spontaneous single- and multi-unit activity in cultured neocortical slices.6 8 And in fact low concentrations of halothane, isoflurane, enflurane, sevoflurane, propofol, etomidate, ethanol, pentobarbital, and ketamine significantly decreased the mean firing rates. In the case of halothane, isoflurane, and enflurane half-maximal depression of action potential firing in brain slices was observed close to the corresponding values reported in vivo. With all anaesthetics, except for halothane, the concentrations causing half-maximal depression of action potential firing were low compared with the EC50 for skin incision.
By what molecular mechanisms do general anaesthetics depress spontaneous spiking in cultured neocortical brain slices? At concentrations causing a half-maximal decrease in the average firing rates most of the tested anaesthetics (except ketamine and pentobarbital) enhanced GABAA receptor-function in a rather selective manner. In summary, it seems that the depressant effects of general anaesthetics on cortical metabolism, blood flow, and spontaneous spiking observed in vivo can be largely attributed to direct actions on cortical neurones. Studies on cultured neocortical brain slices also strengthen the hypothesis that GABAA receptors are major players in mediating the latter effects. This conclusion is consistent with the central role of GABAA receptors in the generation and pharmacological modulation of gamma and theta oscillations discussed above.
Conclusions
General anaesthetics alter the firing of cortical neurones in complex ways. Dependent on the range of concentration, quite different patterns emerge. In this article, two major questions were addressed. First, which of the effects seen in EEG/LFP recordings and imaging studies in vivo can be attributed to drug actions taking place in the cerebral cortex? Secondly, what are the molecular mechanisms involved? For a long time it had been thought that most of the activity patterns of cortical neurones seen under anaesthesia result from effects on sub-cortical structures, thus inducing a sleep-like state. However, recent data, mainly based on brain slice studies, support a different view. Cortical networks turned out to be quite sensitive to anaesthetic treatment. Furthermore, anaesthetic actions in isolated brain slices seem to account for many of the patterns seen also in vivo, thus challenging the hypothesis that sub-cortical actions are most important. From recent findings, it is very likely that the direct depression of cortical neurones largely contributes to hypnosis and amnesia. However, we should not jump to conclusions. The data discussed so far do not exclude the possibility that a brain state similar to natural sleep is caused by many anaesthetics. For example, the finding that a slowing of theta oscillations towards the delta range, as observed in isolated neocortical brain slices, does not argue against the possibility that anaesthetic effects on thalamic neurones such as those schematically shown in Figure 2 are of central importance in producing the low-frequency high-voltage EEG. Unfortunately, there is not much known about general anaesthetic actions on brain centres involved in arousal and sleep regulation. During anaesthesia, two things probably come together: the induction of a sleep-like state by drug actions taking place on a sub-cortical level and, simultaneously, a massive depression of cortical excitability. Which patterns of network activity induced by anaesthetic treatment in vivo cannot be explained by drug actions observed in isolated cortical slices? At least two phenomena are worthy of mention. EEG-activation and spindling do occur at very light levels of anaesthesia. EEG-activation is characterized by an increase in power in all frequency bands. This type of network activity probably depends on sub-cortical structures.63 64 Spindles can be described as oscillations exhibiting waxing and waning amplitudes. They have not been reported to occur in isolated cortical slices. In contrast, such patterns have been investigated in brain slices containing parts of the thalamus and the reticular nucleus.12 70 Most likely, spindles are generated by a complex interplay between thalamic relay neurones and reticular neurones.
Another important result of brain slice studies concerns the molecular mechanisms of anaesthetic actions. It has been known for a long time that GABAA receptors are a quite sensitive molecular target of many anaesthetic agents. The findings that molecular actions at the GABAA receptor are probably causally related to the slowing of synchronized network activity in the gamma and theta band and the depression of mean discharge rates as well, provides an important link between drug actions observed on the molecular and network level.
Even though the data discussed so far seem to fit well together, we should be careful as only a very few parts of the puzzle are known. Comparative studies on anaesthetic effects are recommended. For example, it has still to be shown that anaesthetic effects on cortical gamma oscillations are quantitatively similar in vivo and in vitro. Furthermore, we should be careful in generalizing. In this article, I have strongly emphasized those drug actions many anaesthetics seem to have in common, disregarding agent-specific aspects. This approach was chosen in order to develop some type of working hypothesis capable of integrating much of our current knowledge. However, things may well turn out to be more complex.
Acknowledgements
I thank Kuno Kirschfeld for introducing me into the field of anaesthesiology research and for generous support. Many thanks to Ina Pappe, for her patient help and advice. Finally, thanks to Harald Hentschke for discussing all the topics addressed in this article.
References
1 Alkire MT. Quantitative EEG correlations with brain glucose metabolic rate during anesthesia in volunteers. Anesthesiology 1998; 89: 32333[ISI][Medline]
2 Alkire MT, Haier RJ, Barker SJ, et al. Cerebral metabolism during propofol anesthesia in humans studied with positron emission tomography. Anesthesiology 1995; 82: 393403[ISI][Medline]
3 Alkire MT, Haier RJ, Fallon JH. Toward a unified theory of narcosis: brain imaging evidence for a thalamocortical switch as the neurophysiologic basis of anesthetic-induced uncon sciousness. Conscious Cogn 2000; 9: 37086[ISI][Medline]
4 Alkire MT, Haier RJ, Shah NK, et al. Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia. Anesthesiology 1997; 86: 54957[ISI][Medline]
5 Alkire MT, Pomfrett CJD, Haier RJ, et al. Functional brain imaging during anesthesia in humans. Effects of halothane on global and regional glucose metabolism. Anesthesiology 1999; 90: 7019[ISI][Medline]
6 Antkowiak B. Different actions of general anaesthetics on the firing patterns of neocortical neurons mediated by the GABAA receptor. Anesthesiology 1999; 91: 50011[ISI][Medline]
7 Antkowiak B. The molecular basis of general anaesthesia. Naturwissenschaften 2000; 88: 20113[ISI]
8 Antkowiak B, Helfrich-Förster C. Effects of small concentrations of volatile anesthetics on action potential firing of neocortical neurons in vitro. Anesthesiology 1998; 88: 1592602[ISI][Medline]
9 Antkowiak B, Hentschke H. Cellular mechanisms of gamma rhythms in rat neocortical brain slices probed by the volatile anaesthetic isoflurane. Neurosci Lett 1997; 231: 8790[ISI][Medline]
10 Antognini JF, Carstens E. Isoflurane blunts electro encephalographic and thalamic-reticular formation responses to noxious stimulation in goats. Anesthesiology 1999; 91: 17709[ISI][Medline]
11 Azouz R, Gray CM. Cellular mechanisms contributing to response variability of cortical neurons in vivo. J Neurosci 1999; 19: 220923
12 Bal T, Debay D, Destexhe A. Cortical feedback controls the frequency and synchrony of oscillations in the visual thalamus. J Neurosci 2000; 20: 747888
13 Banks MI, Li T-B, Pearce RA. The synaptic basis of GABAA,slow. J Neurosci 1998; 18: 130517
14 Banks MI, Pearce RA. Kinetic differences between synaptic and extrasynaptic GABAA receptors in CA1 pyramidal cells. J Neurosci 2000; 20: 93748
15 Banks MI, White JA, Pearce RA. Interactions between distinct GABAA circuits in hippocampus. Neuron 2000; 25: 44957[ISI][Medline]
16 Belelli D, Pistis M, Peters JA, et al. General anaesthetic action at transmitter-gated inhibitory amino acid receptors. Trends Pharmacol Sci 1999; 20: 496502[ISI][Medline]
17 Berger H. Über das Elektroenkephalogramm des Menschen. Arch Psychiatr Nervenkr 1929; 87: 52770
18 Blanton MG, Lo Turco JJ, Kriegstein AR. Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex. J Neurosci Methods 1989; 30: 210310
19 Cariani P. Anesthesia, neural information processing, and conscious awareness. Conscious Cogn 2001; 9: 38795[ISI]
20 Clark DL, Rosner BS. Neurophysiologic effects on general anesthetics: I. The electroencephalogram and sensory evoked responses in man. Anesthesiology 1973; 38: 56482[ISI][Medline]
21 Detsch O, Vahle-Hinz C, Kochs E, et al. Isoflurane induces dose-dependent changes of thalamic somatosensory information transfer. Brain Res 1999; 829: 7789[ISI][Medline]
22 Dodt HU, Zieglgänsberger W. Infrared videomicroscopy: a new look at neuronal structure and function. Trends Neurosci 1994; 17: 4538[ISI][Medline]
23 Engel AK, Roelfsema PR, Fries P, et al. Role of the temporal domain for response selection and perceptual binding. Cereb Cortex 1997; 7: 57182[Abstract]
24 Fiset P, Paus T, Daloze T, et al. Brain mechanisms of propofol-induced loss of consciousness in humans: a positron emission tomographic study. J Neurosci 1999; 19: 550613
25 Franks NP, Lieb WR. Molecular and cellular mechanisms of general anaesthesia. Nature 1994; 367: 60714[ISI][Medline]
26 Gaese BH, Ostwald J. Anesthesia changes frequency tuning of neurons in the rat primary auditory cortex. J Neurophysiol 2001; 86: 10626
27 Galarreta M, Hestrin S. Electrical synapses between GABA-releasing interneurons. Nat Rev Neurosci 2001; 2: 42533[ISI][Medline]
28 Gähwiler BH. Organotypic monolayer cultures of nervous tissue. J Neurosci Methods 1981; 4: 32942[ISI][Medline]
29 Gähwiler BH, Capogna M, Debanne D, et al. Organotypic slice cultures: a technique has come of age. Trends Neurosci 1997; 20: 4717[ISI][Medline]
30 Ghose GM, Maunsell J. Specialized representations in visual cortex: a role for binding? Neuron 1999; 24: 7985[ISI][Medline]
31 Gray CM. The temporal correlation hypothesis of visual feature integration: still alive and well. Neuron 1999; 24: 3147[ISI][Medline]
32 Gray CM, McCormick DA. Chattering cells: superficial pyramidal neurons contributing to the generation of synchronous oscillations in the visual cortex. Science 1996; 274: 10913
33 Harris RA, Mihic SJ, Dildy-Mayfield JE, et al. Actions of anesthetics on ligand-gated ion channels: role of receptor subunit composition. FASEB J 1995; 9: 145462
34 Henry CE, Scoville WB. Supression-burst activity from isolated cerebral cortex in man. Electroenceph Clin Neurophyiol 1952; 4: 122
35 Hentschke H, Antkowiak B. NMDA receptor-mediated changes of spontaneous activity patterns in thalamocortical slice cultures. Brain Res 1999; 830: 12337[ISI][Medline]
36 Jefferys JGR, Traub RD, Whittington MA. Neuronal networks for induced 40 Hz rhythms. Trends Neurosci 1996; 19: 2028[ISI][Medline]
37 Keifer JC, Baghdoyan HA, Lydic R. Pontine cholinergic mechanisms modulate the cortical electroencephalographic spindles of halothane anesthesia. Anesthesiology 1996; 84: 94554[ISI][Medline]
38 Leung LS. Nonlinear feedback model of neuronal populations in hippocampal CAl region. J Neurophysiol 1982; 47: 84568
39 Leung LS. Fast (beta) rhythms in the hippocampus: a review. Hippocampus 1992; 2: 938[ISI][Medline]
40 Li CH, McIlwain H. Maintenance of resting membrane potentials in slices of mammalian cerebral cortex and other tissues in vitro. J Physiol 1957; 139: 17890[ISI]
41 Lukatch HS, MacIver MB. Synaptic mechanisms of thiopental-induced alterations in synchronized cortical activity. Anesthesiology 1996; 84: 142534[ISI][Medline]
42 MacIver M, Mandema JW, Stanski DR, et al. Thiopental uncouples hippocampal and cortical synchronized electroencephalographic activity. Anesthesiology 1996; 84: 141124[ISI][Medline]
43 Madler C, Keller I, Schwender D, et al. Sensory information processing during general anaesthesia: effect of isoflurane on auditory evoked neuronal oscillations. Br J Anaesth 1991; 66: 817[Abstract]
44 McCormick DA, Bal T. Sleep and arousal: thalamocortical mechanisms. Annu Rev Neurosci 1997; 20: 185215[ISI][Medline]
45 McIlwain H, Buchel L, Cheshire JD. The inorganic phosphate and phosphocreatine of brain especially during metabolism in vitro. Biochem J 1951; 48: 1220[ISI]
46 McIlwain H, Buddle HL. Techniques in tissue metabolism. I. A mechanical chopper. Biochem J 1953; 53: 41220[ISI]
47 McKernan MG, Rosahl TW, Reynolds DS, et al. Sedative but not anxiolytic properties of benzodiazepines are mediated by the GABAA receptor 1 subtype. Nature Neurosci 2000; 3: 58792[ISI][Medline]
48 Moruzzi G, Magoun HW. Brain stem reticular formation and activation of the EEG. Electroenceph Clin Neurophysiol 1949; 1: 45573[ISI]
49 Munglani R, Jones GJ. Altered consciousness: pharmacology and phenomenology. BAP summer meeting, York, July 1991. Sleep and general anaesthesia as altered states of consciousness. J Psychopharmacol 1992; 6: 399409
50 Nagi SH, Cheney DL, Finck AD. Acetylcholine concentrations and turnover in rat brain structures during anesthesia with halothane, enflurane, and ketamine. Anesthesiology 1978; 48: 410[ISI][Medline]
51 Pearce RA. Physiological evidence for two distinct GABAA responses in rat hippocampus. Neuron 1993; 10: 189200[ISI][Medline]
52 Pöppel E. A hierarchical model of temporal perception. Trends Cogn Sci 1997; 1: 5661
53 Pulvermüller F, Birbaumer N, Lutzenberger W, et al. High-frequency brain activity: its possible role in attention, perception and language processing. Prog Neurobiol 1997; 52: 445
54 Raghavachari S, Kahana MJ, Rizzuto DS, et al. Gating of human theta oscillations by a working memory task. J Neurosci 2001; 21: 317583
55 Ries CR, Puil E. Mechanism of anesthesia revealed by shunting actions of isoflurane on thalamocortical neurons. J Neurophysiol 1999; 81: 1795999
56 Roskies AL. The binding problem. Neuron 1999; 24: 79[ISI][Medline]
57 Rudolph U, Crestani F, Benke D, et al. Benzodiazepine actions mediated by specific -aminobutyric acidA receptor subtypes. Nature 1999; 401: 796800[ISI][Medline]
58 Schleidt M, Eibl-Eibesfeld I, Pöppel E. A universal constant in temporal segmentation of human short-term behavior. Naturwissenschaften 1987; 74: 28990[ISI][Medline]
59 Shadlen MN, Movshon JA. Synchrony unbound: a critical evaluation of the temporal binding hypothesis. Neuron 1999; 24: 6777[ISI][Medline]
60 Singer W. Synchronization of cortical activity and its putative role in information processing and learning. Annu Rev Physiol 1993; 55: 34974[ISI][Medline]
61 Singer W. Neuronal synchrony: a versatile code for the definition of relations? Neuron 1999; 24: 4965[ISI][Medline]
62 Singer W, Gray CM. Visual feature integration and the temporal correlation hypothesis. Annu Rev Neurosci 1995; 18: 55586[ISI][Medline]
63 Steriade M. Arousal: revisiting the reticular activating system. Science 1996; 272: 2256[ISI][Medline]
64 Steriade M, McCormick DA, Sejnowski TJ. Thalamocortical oscillations in the sleeping and aroused brain. Science 1993; 262: 67985[ISI][Medline]
65 Stuart GJ, Dodt HU, Sakmann B. Patch-clamp recordings from the soma and dendrites of neurons in brain slices using infrared video microscopy. Pflügers Arch 1993; 423: 5118[ISI][Medline]
66 Swank RL. Synchronization of spontaneous electrical activity of cerebrum by barbiturate narcosis. J Neurophysiol 1949; 12: 16172
67 Tanelian DL, Kosek P, Mody I, et al. The role of the GABAA receptor/chloride channel complex in anesthesia. Anesthesiology 1993; 78: 75776[ISI][Medline]
68 Thornton C. Evoked potentials in anaesthesia. Eur J Anaesth 1991; 8: 89107[ISI][Medline]
69 Uchida S, Nakayama H, Maehara T, et al. Suppression of gamma activity in the human medial temporal lobe by sevoflurane anesthesia. Neuroreport 2000; 11: 3942[ISI][Medline]
70 Vonkrosigk M, Bal T, McCormick DA. Cellular mechanisms of a synchronized oscillation in the thalamus. Science 1993; 261: 3614[ISI][Medline]
71 White AE, Banks MI, Pearce RA, et al. Network of interneurons with fast and slow -aminobutyric acid A (GABAA) kinetics provide substrate for mixed gamma-theta rhythm. Proc Natl Acad Sci USA 2000; 97: 812833
72 Whittington MA, Jefferys JGR, Traub RD. Effects of intravenous anaesthetic agents on fast inhibitory oscillations in the rat hippocampus in vitro. Br J Pharmacolol 1996; 118: 197786[Abstract]
73 Whittington MA, Stanford IM, Colling SB, et al. Spatiotemporal patterns of gamma frequency oscillations tetanically induced in the rat hippocampal slice. J Physiol 1997; 502: 591607[Abstract]
74 Whittington MA, Traub RD, Jefferys JGR. Synchronized oscillations in interneuron networks driven by metabotropic glutamate receptor activation. Nature 1995; 373: 6125[ISI][Medline]
75 Yamamoto C, McIlwain H. Electrical activities in thin sections from the mammalian brain maintained in chemically-defined media in vitro. J Neurochem 1966; 13: 133343[ISI][Medline]