- Email: [email protected]
Increased responsiveness of cortical neurons in contrast to thalamic neurons during isoflurane-induced EEG bursts in rats
Neuroscience Letters 317 (2002) 9–12 www.elsevier.com/locate/neulet
Increased responsiveness of cortical neurons in contrast to thalamic neurons during isoflurane-induced EEG bursts in rats Oliver Detsch a, Eberhard Kochs a, Matthias Siemers b, Burkhart Bromm b, Christiane Vahle-Hinz b,* a
b
Klinik fu¨r Anaesthesiologie, Technische Universita¨t Mu¨nchen, Munich, Germany Institut fu¨r Physiologie, Universita¨tsklinikum Hamburg–Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany Received 11 September 2001; accepted 12 October 2001
Abstract The neuronal mechanisms underlying the electroencephalographic (EEG) burst–suppression pattern are not yet understood, however, they are generally attributed to interactions within thalamocortical networks. In contrast, we report that the sensory cortex and the thalamus are disconnected, with thalamic sensory processing being unaffected by cortical EEG bursts. We studied the activity of single neurons of the somatosensory thalamocortical system in rats during burst–suppression EEG induced by the volatile anesthetic, isoflurane. In neurons of the thalamic ventrobasal complex, the discharge rate in response to tactile stimulation of their receptive fields did not differ significantly during EEG bursts and isoelectric periods. In contrast, in neurons of the primary somatosensory cortex, the response magnitude was significantly greater during EEG bursts as compared with isoelectric periods (mean increase to 293%). The results suggest that the profound suppression of cortical sensory information processing by isoflurane is suspended during EEG burst-induced elevated cortical excitation. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Volatile anesthetics; Isoflurane; Electroencephalogram; Burst–suppression pattern; Ventrobasal complex; Thalamus; Somatosensory cortex
The electroencephalographic (EEG) pattern of burst– suppression consists of alternating periods of high voltage bursts and periods of very low voltage or even isoelectricity. It may be reversibly induced by anesthetics, e.g. the volatile anesthetic, isoflurane [3]. The neuronal mechanisms underlying the EEG burst–suppression pattern are not yet understood. Data are sparse concerning the activity of single neurons corresponding to the EEG burst–suppression pattern, however, it has been put forward that interactions between thalamic and cortical neurons may be associated with this peculiar EEG pattern [15]. The aim of the present study was to analyze the action potential discharge activity of cortical and thalamic neurons during EEG burst versus isoelectric periods in rats anesthetized with isoflurane, particularly with respect to their responses to sensory stimulation. With IACUC (Institutional Animal Care and Use Committee) approval, ten adult Wistar rats were prepared * Corresponding author. Tel.: 149-40-42803-4789; fax: 149-4042803-4920. E-mail address: [email protected] (C. Vahle-Hinz).
for acute electrophysiological recordings with methods described previously [5]. Anesthesia was induced and maintained exclusively with isoflurane. The rats were mechanically ventilated, and end-tidal pCO2 as well as rectal temperature were maintained in normal ranges. The mean arterial blood pressure was $80 mmHg in all rats throughout the study period without the use of vasopressors. After the experiments, the rats were killed with pentobarbital. To induce the EEG burst–suppression pattern, end-tidal isoflurane concentrations between 1.6 and 2.2 vol.% were used (1.9 ^ 0.04 and 1.9 ^ 0.02 vol.% for thalamic and cortical recordings, respectively; P ¼ 0:52). To quantify the extent of EEG suppression, the electrocorticogram (ECoG) was analyzed using the burst–suppression ratio (BSR; i.e. the percentage of isoelectric periods [12]), the number of bursts/min, and the mean duration of the bursts. A similar ECoG pattern was induced in both series of experiments, i.e. during recordings of thalamic and cortical neurons (BSR, 0.73 ^ 0.04 and 0.67 ^ 0.02, P ¼ 0:14; number of bursts/ min, 9.4 ^ 1.0 and 10.7 ^ 0.7, P ¼ 0:27; burst duration, 1.8 ^ 0.2 and 1.9 ^ 0.1 s, P ¼ 0:53). For ECoG recording, a silver wire was implanted through
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 41 9- 3
10
O. Detsch et al. / Neuroscience Letters 317 (2002) 9–12
a burr hole contacting the dura mater above the barrel field of the left primary somatosensory cortex (S1). The ECoG was recorded against an ear bar of the stereotaxic apparatus as reference, filtered at 0.1 Hz and 3 kHz, and stored on a computer with Spike2 software (sampling rate, 10 kHz). For extracellular recordings of single neurons, a tungsten electrode was inserted stereotaxically into the right ventrobasal complex (VB) or lamina IV/V of the right S1. The cortical recording sites were determined by electrolytic lesions and histological verification; the thalamic recording sites were ascertained electrophysiologically [5,16]. Mechanosensory neurons in VB and S1 responding with sustained discharges to vibration of whiskers under ,1.0 vol.% isoflurane were selected. Sinusoidal shaped stimuli (20–600 Hz; duration, 0.5 s; repetition rate, 0.4/s) were applied to single whiskers via a feedback-controlled electromechanical stimulator. Response activities of VB and S1 neurons were assessed separately for their rates during EEG bursts and during isoelectric phases for periods of 2 min. The response activity per stimulus was determined from peristimulus time histograms (bin width, 1 ms) as the mean discharge frequency (spikes/s) above ongoing activity; ongoing activity was measured from the 1-s period immediately before stimulus onset, or from a 1-s period of the preceding burst or isoelectric phase. The data were pooled for statistical comparisons, and where appropriate, paired as well as unpaired Student’s t-tests were used, with P , 0:05
considered as significant. Data are reported as means ^ SEM. Data were derived from 18 S1 neurons and 21 VB neurons. Fig. 1A shows sample records of the burst– suppression ECoG and the discharge activities of a S1 neuron (see also Fig. 2A) and a VB neuron elicited by vibration of whiskers within their respective receptive fields. The response activity of VB neurons did not differ significantly during ECoG isoelectricity and bursts (Fig. 1A,B). In contrast, the response activity of S1 neurons was 1.2 ^ 0.3 spikes/s during ECoG isoelectricity and increased significantly to 3.5 ^ 0.6 spikes/s during bursts (P , 0:001) representing a mean increase to 293%. Thereby, during EEG bursts, the responsiveness of S1 neurons was similar to that of VB neurons during both EEG phases. The ongoing activity of VB neurons did not differ between both EEG phases and was not different from ongoing activity of S1 neurons during EEG isoelectricity. During bursts, S1 ongoing activity increased markedly (P , 0:001; Figs. 1B and 2B). The major findings of the present study during isofluraneinduced burst–suppression ECoG are: (i), the discharges of cortical neurons in response to stimulation of their receptive fields were significantly higher during bursts as compared with isoelectric periods; (ii), the response magnitude of subcortical neurons was not affected by the ECoG periods; and (iii), during ECoG bursts, the responsiveness of cortical
Fig. 1. Burst–suppression ECoG induced by isoflurane and neuronal action potential discharges. (A) The ECoG was recorded from the S1 contralateral to the recording sites of a single neuron in S1 (left) and the thalamic VB (right). In contrast to the VB neuron, the S1 neuron responded with sustained action potential discharges (spike events) to vibration of a whisker (stimulus trace) only during an ECoG burst. (B) The response activity of S1 neurons (n ¼ 18) was significantly higher during EEG bursts than during isoelectric periods. In contrast, the response activity of VB neurons (n ¼ 21) was not affected by the EEG phases. Note that the cortical response activity during bursts reached the magnitude of the thalamic response activity. The response activity was corrected for ongoing activity which also increased in S1 neurons during bursts (right). Means 1 SEM, *P , 0:001.
O. Detsch et al. / Neuroscience Letters 317 (2002) 9–12
11
Fig. 2. (A) The occurrence of electrocorticographic (ECoG) bursts was independent of sensory stimulation. ECoG bursts were not associated with vibratory stimuli (110 Hz, 0.5 s; stimulus trace) applied to the whisker representing the neuron’s receptive field. (B) The ongoing activity (spike events) of the neuron of the S1 was correlated with ECoG bursts (no stimulation of the neuron’s receptive field). Recordings of the same cortical neuron in (A) and (B).
neurons reached the response magnitude of thalamocortical relay (TC) neurons, which constitute their afferent sensory input. The results suggest that during EEG burst-induced elevated excitation, cortical sensory suppression is suspended, although cortical information processing is profoundly suppressed by high concentrations of isoflurane. This effect appears to be independent of the activity of the sensory thalamus and, vice versa, does not affect thalamic sensory processing. Our results cannot be explained by differences in the degree of EEG suppression for the cortical and thalamic neuronal populations, since the quantitative EEG parameters did not differ, and moreover, the isoflurane concentrations used to induce the burst–suppression EEG were equal for cortical and thalamic experiments. It was hypothesized that periods of EEG suppression may be produced by an inhibitory mechanism acting at the cortex, and that short periods of excitation are superimposed by a non-linear EEG generator resulting in the burst– suppression pattern of EEG [10,18]. However, the source of the input causing excitatory periods (i.e. bursts) remains speculative. It was suggested that ascending inputs may constitute such a source, since bursts at cortical and subcortical sites may be highly synchronized [1,9]. Moreover, complex actions of cortico-thalamo-cortical loops have been shown to trigger and synchronize thalamic oscillations [2,4,11], and thus, it is conceivable that these interactions between cortical and thalamic mechanisms may also be responsible for the bursts during the hypersynchronized state of burst–suppression EEG. Though sensory stimulation (such as visual, auditory and somatosensory stimuli) was shown to induce EEG bursts in humans [7,18], our data showed that there was no association between periods
of stimulation and periods of EEG bursts (compare Fig. 2A,B). Thus, we have no indication that stimulus-evoked thalamic activity, at least such as that evoked by vibratory stimuli applied to single whiskers, is a source for cortical bursts. On the contrary, the independence of cortical bursts from the direct afferent inputs from VB is evident. The present results are in accordance with a previous study [8], in which isolated neocortical brain slices (i.e. cortex preparations without a subcortical input) exerted some burst–suppression activity, suggesting a mechanism intrinsic to the cerebral cortex. Also, Steriade et al. [15], investigating the cellular correlates of the burst–suppression EEG, found a markedly reduced responsiveness of cortical neurons to synaptic volleys induced by electrical stimulation during EEG isoelectricity. This is in agreement with our findings using natural stimulation of the receptive fields. Also, the close relationship between the activity of cortical neurons recorded intracellularly and EEG activity [15] is in line with our finding of increased ongoing action potential discharges of S1 neurons during EEG bursts. At variance with cortical neurons, thalamic neurons were reported to still exert activity while EEG activity was suppressed [15]. For instance, the ongoing and evoked activities of a TC neuron of the visual thalamus did not change when EEG bursts occurred [15], which corroborates the results from our recordings of the somatosensory thalamus. During cortically generated seizures, or other types of sleep oscillation displayed by cortical neurons before entering burst–suppression EEG, the highly synchronized cortical activity results in powerful cycles of phasic inhibition of the thalamus mediated through the thalamic reticular nucleus, which results in burst-firing of thalamic neurons
12
O. Detsch et al. / Neuroscience Letters 317 (2002) 9–12
[2,4,11,14]. This inhibition overrides the excitatory projection from the cortex to TC neurons (seen only during singlespike modes) since the synaptic strength of the excitatory cortical inputs to neurons of the thalamic reticular nucleus is greater than that to TC neurons [6,13]. Since isoflurane anesthesia has been shown to further enhance thalamic inhibition [17], it was expected that VB neurons would display rhythmic ongoing activity and stimulus-elicited responses would be suppressed during EEG bursts. However, the results showed neither of these effects, but thalamic activity independent of EEG phases and without signs of cellular hyperpolarization. With respect to processing of sensory information under deep levels of anesthesia, it is interesting to note that during EEG bursts, information about tactile stimuli is still encoded by cortical neurons. [1] Akrawi, W.P., Drummond, J.C., Kalkman, C.J. and Patel, P.M., A comparison of the electrophysiologic characteristics of EEG burst–suppression as produced by isoflurane, thiopental, etomidate, and propofol, J. Neurosurg. Anesth., 8 (1996) 40–46. [2] Blumenfeld, H. and McCormick, D.A., Corticothalamic inputs control the pattern of activity generated in thalamocortical networks, J. Neurosci., 20 (2000) 5153–5162. [3] Clark, D.L., Hosick, E.C., Adam, N., Castro, A.D., Rosner, B.S. and Neigh, J.L., Neural effects of isoflurane (forane) in man, Anesthesiology, 39 (1973) 261–270. [4] Destexhe, A., Contreras, D. and Steriade, M., Corticallyinduced coherence of a thalamic-generated oscillation, Neuroscience, 92 (1999) 427–443. [5] Detsch, O., Vahle-Hinz, C., Kochs, E., Siemers, M. and Bromm, B., Isoflurane induces dose-dependent changes of thalamic somatosensory information transfer, Brain Res., 829 (1999) 77–89. [6] Golshani, P., Liu, X.-B. and Jones, E., Differences in quantal amplitude reflect GluR4-subunit number at corticothalamic synapses on two populations of thalamic neurons, Proc. Natl. Acad. Sci. USA, 98 (2001) 4172–4177. [7] Hartikainen, K.M., Rorarius, M., Perakyla, J.J., Laippala, P.J. and Jantti, V., Cortical reactivity during isoflurane burst–
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
suppression anesthesia, Anesth. Analg., 81 (1995) 1223– 1228. Lukatch, H.S. and MacIver, M.B., Synaptic mechanisms of thiopental-induced alterations in synchronized cortical activity, Anesthesiology, 84 (1996) 1425–1434. MacIver, M.B., Mandema, J.W., Stanski, D.R. and Bland, B.H., Thiopental uncouples hippocampal and cortical synchronized electroencephalographic activity, Anesthesiology, 84 (1996) 1411–1424. Muthuswamy, J., Sherman, D.L. and Thakor, N.V., Higherorder spectral analysis of burst patterns in EEG, IEEE Trans. Biomed. Eng., 46 (1999) 92–99. Neckelmann, D., Amzica, F. and Steriade, M., Spike-wave complexes and fast components of cortically generated seizures. III. Synchronizing mechanisms, J. Neurophysiol., 80 (1998) 1480–1494. Rampil, I.J., Weiskopf, R.B., Brown, J.G., Eger 2nd, E.I., Johnson, B.H., Holmes, M.A. and Donegan, J.H., I653 and isoflurane produce similar dose-related changes in the electroencephalogram of pigs, Anesthesiology, 69 (1988) 298–302. Steriade, M., The GABAergic reticular nucleus: a preferential target of corticothalamic projections, Proc. Natl. Acad. Sci. USA, 98 (2001) 3625–3627. Steriade, M. and Contreras, D., Relations between cortical and thalamic cellular events during transition from sleep patterns to paroxysmal activity, J. Neurosci., 15 (1995) 623–642. Steriade, M., Amzica, F. and Contreras, D., Cortical and thalamic cellular correlates of electroencephalographic burst–suppression, Electroenceph. clin. Neurophysiol., 90 (1994) 1–16. Vahle-Hinz, C. and Gottschaldt, K.M., Principal differences in the organization of the thalamic face representation in rodents and felids, In G. Macchi, A. Rustioni and R. Spreafico (Eds.), Somatosensory Integration in the Thalamus, Elsevier, Amsterdam, 1983, pp. 125–145. Vahle-Hinz, C., Detsch, O., Siemers, M., Kochs, E. and Bromm, B., Local GABAA receptor blockade reverses isoflurane’s suppressive effects on thalamic neurons in vivo, Anesth. Analg., 92 (2001) 1578–1584. Yli-Hankala, A., Ja¨ntti, V., Pyykko¨, I. and Lindgren, L., Vibration stimulus induced EEG bursts in isoflurane anaesthesia, Electroenceph. clin. Neurophysiol., 87 (1993) 215–220.
Increased responsiveness of cortical neurons in contrast to thalamic neurons during isoflurane-induced EEG bursts in rats Oliver Detsch a, Eberhard Kochs a, Matthias Siemers b, Burkhart Bromm b, Christiane Vahle-Hinz b,* a
b
Klinik fu¨r Anaesthesiologie, Technische Universita¨t Mu¨nchen, Munich, Germany Institut fu¨r Physiologie, Universita¨tsklinikum Hamburg–Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany Received 11 September 2001; accepted 12 October 2001
Abstract The neuronal mechanisms underlying the electroencephalographic (EEG) burst–suppression pattern are not yet understood, however, they are generally attributed to interactions within thalamocortical networks. In contrast, we report that the sensory cortex and the thalamus are disconnected, with thalamic sensory processing being unaffected by cortical EEG bursts. We studied the activity of single neurons of the somatosensory thalamocortical system in rats during burst–suppression EEG induced by the volatile anesthetic, isoflurane. In neurons of the thalamic ventrobasal complex, the discharge rate in response to tactile stimulation of their receptive fields did not differ significantly during EEG bursts and isoelectric periods. In contrast, in neurons of the primary somatosensory cortex, the response magnitude was significantly greater during EEG bursts as compared with isoelectric periods (mean increase to 293%). The results suggest that the profound suppression of cortical sensory information processing by isoflurane is suspended during EEG burst-induced elevated cortical excitation. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Volatile anesthetics; Isoflurane; Electroencephalogram; Burst–suppression pattern; Ventrobasal complex; Thalamus; Somatosensory cortex
The electroencephalographic (EEG) pattern of burst– suppression consists of alternating periods of high voltage bursts and periods of very low voltage or even isoelectricity. It may be reversibly induced by anesthetics, e.g. the volatile anesthetic, isoflurane [3]. The neuronal mechanisms underlying the EEG burst–suppression pattern are not yet understood. Data are sparse concerning the activity of single neurons corresponding to the EEG burst–suppression pattern, however, it has been put forward that interactions between thalamic and cortical neurons may be associated with this peculiar EEG pattern [15]. The aim of the present study was to analyze the action potential discharge activity of cortical and thalamic neurons during EEG burst versus isoelectric periods in rats anesthetized with isoflurane, particularly with respect to their responses to sensory stimulation. With IACUC (Institutional Animal Care and Use Committee) approval, ten adult Wistar rats were prepared * Corresponding author. Tel.: 149-40-42803-4789; fax: 149-4042803-4920. E-mail address: [email protected] (C. Vahle-Hinz).
for acute electrophysiological recordings with methods described previously [5]. Anesthesia was induced and maintained exclusively with isoflurane. The rats were mechanically ventilated, and end-tidal pCO2 as well as rectal temperature were maintained in normal ranges. The mean arterial blood pressure was $80 mmHg in all rats throughout the study period without the use of vasopressors. After the experiments, the rats were killed with pentobarbital. To induce the EEG burst–suppression pattern, end-tidal isoflurane concentrations between 1.6 and 2.2 vol.% were used (1.9 ^ 0.04 and 1.9 ^ 0.02 vol.% for thalamic and cortical recordings, respectively; P ¼ 0:52). To quantify the extent of EEG suppression, the electrocorticogram (ECoG) was analyzed using the burst–suppression ratio (BSR; i.e. the percentage of isoelectric periods [12]), the number of bursts/min, and the mean duration of the bursts. A similar ECoG pattern was induced in both series of experiments, i.e. during recordings of thalamic and cortical neurons (BSR, 0.73 ^ 0.04 and 0.67 ^ 0.02, P ¼ 0:14; number of bursts/ min, 9.4 ^ 1.0 and 10.7 ^ 0.7, P ¼ 0:27; burst duration, 1.8 ^ 0.2 and 1.9 ^ 0.1 s, P ¼ 0:53). For ECoG recording, a silver wire was implanted through
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 41 9- 3
10
O. Detsch et al. / Neuroscience Letters 317 (2002) 9–12
a burr hole contacting the dura mater above the barrel field of the left primary somatosensory cortex (S1). The ECoG was recorded against an ear bar of the stereotaxic apparatus as reference, filtered at 0.1 Hz and 3 kHz, and stored on a computer with Spike2 software (sampling rate, 10 kHz). For extracellular recordings of single neurons, a tungsten electrode was inserted stereotaxically into the right ventrobasal complex (VB) or lamina IV/V of the right S1. The cortical recording sites were determined by electrolytic lesions and histological verification; the thalamic recording sites were ascertained electrophysiologically [5,16]. Mechanosensory neurons in VB and S1 responding with sustained discharges to vibration of whiskers under ,1.0 vol.% isoflurane were selected. Sinusoidal shaped stimuli (20–600 Hz; duration, 0.5 s; repetition rate, 0.4/s) were applied to single whiskers via a feedback-controlled electromechanical stimulator. Response activities of VB and S1 neurons were assessed separately for their rates during EEG bursts and during isoelectric phases for periods of 2 min. The response activity per stimulus was determined from peristimulus time histograms (bin width, 1 ms) as the mean discharge frequency (spikes/s) above ongoing activity; ongoing activity was measured from the 1-s period immediately before stimulus onset, or from a 1-s period of the preceding burst or isoelectric phase. The data were pooled for statistical comparisons, and where appropriate, paired as well as unpaired Student’s t-tests were used, with P , 0:05
considered as significant. Data are reported as means ^ SEM. Data were derived from 18 S1 neurons and 21 VB neurons. Fig. 1A shows sample records of the burst– suppression ECoG and the discharge activities of a S1 neuron (see also Fig. 2A) and a VB neuron elicited by vibration of whiskers within their respective receptive fields. The response activity of VB neurons did not differ significantly during ECoG isoelectricity and bursts (Fig. 1A,B). In contrast, the response activity of S1 neurons was 1.2 ^ 0.3 spikes/s during ECoG isoelectricity and increased significantly to 3.5 ^ 0.6 spikes/s during bursts (P , 0:001) representing a mean increase to 293%. Thereby, during EEG bursts, the responsiveness of S1 neurons was similar to that of VB neurons during both EEG phases. The ongoing activity of VB neurons did not differ between both EEG phases and was not different from ongoing activity of S1 neurons during EEG isoelectricity. During bursts, S1 ongoing activity increased markedly (P , 0:001; Figs. 1B and 2B). The major findings of the present study during isofluraneinduced burst–suppression ECoG are: (i), the discharges of cortical neurons in response to stimulation of their receptive fields were significantly higher during bursts as compared with isoelectric periods; (ii), the response magnitude of subcortical neurons was not affected by the ECoG periods; and (iii), during ECoG bursts, the responsiveness of cortical
Fig. 1. Burst–suppression ECoG induced by isoflurane and neuronal action potential discharges. (A) The ECoG was recorded from the S1 contralateral to the recording sites of a single neuron in S1 (left) and the thalamic VB (right). In contrast to the VB neuron, the S1 neuron responded with sustained action potential discharges (spike events) to vibration of a whisker (stimulus trace) only during an ECoG burst. (B) The response activity of S1 neurons (n ¼ 18) was significantly higher during EEG bursts than during isoelectric periods. In contrast, the response activity of VB neurons (n ¼ 21) was not affected by the EEG phases. Note that the cortical response activity during bursts reached the magnitude of the thalamic response activity. The response activity was corrected for ongoing activity which also increased in S1 neurons during bursts (right). Means 1 SEM, *P , 0:001.
O. Detsch et al. / Neuroscience Letters 317 (2002) 9–12
11
Fig. 2. (A) The occurrence of electrocorticographic (ECoG) bursts was independent of sensory stimulation. ECoG bursts were not associated with vibratory stimuli (110 Hz, 0.5 s; stimulus trace) applied to the whisker representing the neuron’s receptive field. (B) The ongoing activity (spike events) of the neuron of the S1 was correlated with ECoG bursts (no stimulation of the neuron’s receptive field). Recordings of the same cortical neuron in (A) and (B).
neurons reached the response magnitude of thalamocortical relay (TC) neurons, which constitute their afferent sensory input. The results suggest that during EEG burst-induced elevated excitation, cortical sensory suppression is suspended, although cortical information processing is profoundly suppressed by high concentrations of isoflurane. This effect appears to be independent of the activity of the sensory thalamus and, vice versa, does not affect thalamic sensory processing. Our results cannot be explained by differences in the degree of EEG suppression for the cortical and thalamic neuronal populations, since the quantitative EEG parameters did not differ, and moreover, the isoflurane concentrations used to induce the burst–suppression EEG were equal for cortical and thalamic experiments. It was hypothesized that periods of EEG suppression may be produced by an inhibitory mechanism acting at the cortex, and that short periods of excitation are superimposed by a non-linear EEG generator resulting in the burst– suppression pattern of EEG [10,18]. However, the source of the input causing excitatory periods (i.e. bursts) remains speculative. It was suggested that ascending inputs may constitute such a source, since bursts at cortical and subcortical sites may be highly synchronized [1,9]. Moreover, complex actions of cortico-thalamo-cortical loops have been shown to trigger and synchronize thalamic oscillations [2,4,11], and thus, it is conceivable that these interactions between cortical and thalamic mechanisms may also be responsible for the bursts during the hypersynchronized state of burst–suppression EEG. Though sensory stimulation (such as visual, auditory and somatosensory stimuli) was shown to induce EEG bursts in humans [7,18], our data showed that there was no association between periods
of stimulation and periods of EEG bursts (compare Fig. 2A,B). Thus, we have no indication that stimulus-evoked thalamic activity, at least such as that evoked by vibratory stimuli applied to single whiskers, is a source for cortical bursts. On the contrary, the independence of cortical bursts from the direct afferent inputs from VB is evident. The present results are in accordance with a previous study [8], in which isolated neocortical brain slices (i.e. cortex preparations without a subcortical input) exerted some burst–suppression activity, suggesting a mechanism intrinsic to the cerebral cortex. Also, Steriade et al. [15], investigating the cellular correlates of the burst–suppression EEG, found a markedly reduced responsiveness of cortical neurons to synaptic volleys induced by electrical stimulation during EEG isoelectricity. This is in agreement with our findings using natural stimulation of the receptive fields. Also, the close relationship between the activity of cortical neurons recorded intracellularly and EEG activity [15] is in line with our finding of increased ongoing action potential discharges of S1 neurons during EEG bursts. At variance with cortical neurons, thalamic neurons were reported to still exert activity while EEG activity was suppressed [15]. For instance, the ongoing and evoked activities of a TC neuron of the visual thalamus did not change when EEG bursts occurred [15], which corroborates the results from our recordings of the somatosensory thalamus. During cortically generated seizures, or other types of sleep oscillation displayed by cortical neurons before entering burst–suppression EEG, the highly synchronized cortical activity results in powerful cycles of phasic inhibition of the thalamus mediated through the thalamic reticular nucleus, which results in burst-firing of thalamic neurons
12
O. Detsch et al. / Neuroscience Letters 317 (2002) 9–12
[2,4,11,14]. This inhibition overrides the excitatory projection from the cortex to TC neurons (seen only during singlespike modes) since the synaptic strength of the excitatory cortical inputs to neurons of the thalamic reticular nucleus is greater than that to TC neurons [6,13]. Since isoflurane anesthesia has been shown to further enhance thalamic inhibition [17], it was expected that VB neurons would display rhythmic ongoing activity and stimulus-elicited responses would be suppressed during EEG bursts. However, the results showed neither of these effects, but thalamic activity independent of EEG phases and without signs of cellular hyperpolarization. With respect to processing of sensory information under deep levels of anesthesia, it is interesting to note that during EEG bursts, information about tactile stimuli is still encoded by cortical neurons. [1] Akrawi, W.P., Drummond, J.C., Kalkman, C.J. and Patel, P.M., A comparison of the electrophysiologic characteristics of EEG burst–suppression as produced by isoflurane, thiopental, etomidate, and propofol, J. Neurosurg. Anesth., 8 (1996) 40–46. [2] Blumenfeld, H. and McCormick, D.A., Corticothalamic inputs control the pattern of activity generated in thalamocortical networks, J. Neurosci., 20 (2000) 5153–5162. [3] Clark, D.L., Hosick, E.C., Adam, N., Castro, A.D., Rosner, B.S. and Neigh, J.L., Neural effects of isoflurane (forane) in man, Anesthesiology, 39 (1973) 261–270. [4] Destexhe, A., Contreras, D. and Steriade, M., Corticallyinduced coherence of a thalamic-generated oscillation, Neuroscience, 92 (1999) 427–443. [5] Detsch, O., Vahle-Hinz, C., Kochs, E., Siemers, M. and Bromm, B., Isoflurane induces dose-dependent changes of thalamic somatosensory information transfer, Brain Res., 829 (1999) 77–89. [6] Golshani, P., Liu, X.-B. and Jones, E., Differences in quantal amplitude reflect GluR4-subunit number at corticothalamic synapses on two populations of thalamic neurons, Proc. Natl. Acad. Sci. USA, 98 (2001) 4172–4177. [7] Hartikainen, K.M., Rorarius, M., Perakyla, J.J., Laippala, P.J. and Jantti, V., Cortical reactivity during isoflurane burst–
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
suppression anesthesia, Anesth. Analg., 81 (1995) 1223– 1228. Lukatch, H.S. and MacIver, M.B., Synaptic mechanisms of thiopental-induced alterations in synchronized cortical activity, Anesthesiology, 84 (1996) 1425–1434. MacIver, M.B., Mandema, J.W., Stanski, D.R. and Bland, B.H., Thiopental uncouples hippocampal and cortical synchronized electroencephalographic activity, Anesthesiology, 84 (1996) 1411–1424. Muthuswamy, J., Sherman, D.L. and Thakor, N.V., Higherorder spectral analysis of burst patterns in EEG, IEEE Trans. Biomed. Eng., 46 (1999) 92–99. Neckelmann, D., Amzica, F. and Steriade, M., Spike-wave complexes and fast components of cortically generated seizures. III. Synchronizing mechanisms, J. Neurophysiol., 80 (1998) 1480–1494. Rampil, I.J., Weiskopf, R.B., Brown, J.G., Eger 2nd, E.I., Johnson, B.H., Holmes, M.A. and Donegan, J.H., I653 and isoflurane produce similar dose-related changes in the electroencephalogram of pigs, Anesthesiology, 69 (1988) 298–302. Steriade, M., The GABAergic reticular nucleus: a preferential target of corticothalamic projections, Proc. Natl. Acad. Sci. USA, 98 (2001) 3625–3627. Steriade, M. and Contreras, D., Relations between cortical and thalamic cellular events during transition from sleep patterns to paroxysmal activity, J. Neurosci., 15 (1995) 623–642. Steriade, M., Amzica, F. and Contreras, D., Cortical and thalamic cellular correlates of electroencephalographic burst–suppression, Electroenceph. clin. Neurophysiol., 90 (1994) 1–16. Vahle-Hinz, C. and Gottschaldt, K.M., Principal differences in the organization of the thalamic face representation in rodents and felids, In G. Macchi, A. Rustioni and R. Spreafico (Eds.), Somatosensory Integration in the Thalamus, Elsevier, Amsterdam, 1983, pp. 125–145. Vahle-Hinz, C., Detsch, O., Siemers, M., Kochs, E. and Bromm, B., Local GABAA receptor blockade reverses isoflurane’s suppressive effects on thalamic neurons in vivo, Anesth. Analg., 92 (2001) 1578–1584. Yli-Hankala, A., Ja¨ntti, V., Pyykko¨, I. and Lindgren, L., Vibration stimulus induced EEG bursts in isoflurane anaesthesia, Electroenceph. clin. Neurophysiol., 87 (1993) 215–220.