| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Many in vitro effects of volatile anesthetics are known, but the mechanisms of action are still under debate. Because suppression of sensory perception is one of the major goals of general anesthesia, we studied the effects of isoflurane on the processing of somatosensory information in anesthetized rats. Local iontophoretic administration of the -aminobutyric acid-A (GABAA) receptor antagonist bicuculline in the thalamic ventral posteromedial nucleus reversed suppressive effects of isoflurane on thalamocortical relay neurons (TCNs). The action potential discharges of TCNs (n = 23) in response to defined mechanical stimulation of receptive fields seen with small concentrations of isoflurane (0.79% ± 0.01%, mean ± SEM) were suppressed under large concentrations (1.44% ± 0.04%). In addition, the tonic response pattern was lost, which initially encoded the information about the stimulus features. In 70% of TCNs, bicuculline administration reestablished the initially present tonic response pattern under large isoflurane concentrations. These results indicate that isoflurane suppresses somatosensory information transfer at the thalamic level in vivo, apparently by enhancing thalamic GABAA receptor-mediated inhibition.
Implications: Isoflurane actions in the thalamus suppressed the transmission of tactile input to the cortex. This effect was reversed by removal of thalamic inhibition. Suppression of sensory perception under general anesthesia, therefore, may result in part from isofluranes interaction with thalamic
There is controversy about the exact mechanisms of action of volatile anesthetics. Suppression of neuronal electrical activity by volatile anesthetics has been ascribed to interactions with -aminobutyric acid-A (GABAA) receptors (14) and other ligand-gated (3), as well as voltage-gated, ion channels (1). Furthermore, some studies also report that isoflurane (ISO) induces alterations of intrinsic membrane properties of neurons (5,6). It is unclear which of the effects of volatile anesthetics demonstrated in vitro are important for producing relevant in vivo effects seen on the behavioral level. There is increasing evidence from recent experiments with different in vitro brain preparations (79) that the actions of volatile anesthetics are mediated mainly by an enhancement of GABAA-receptor-mediated synaptic inhibition. The relative impact of effects demonstrated in those artificial preparations, however, has to be investigated in fully intact in vivo preparations. We used the trigeminal somatosensory system of the rat as a model. In rats, facial whiskers arranged around the snout drive all types of mechanoreceptors also present in the human skin; therefore, the whisker system has been used extensively as a model for studies of sensory information processing (10). Tactile events are encoded in trains of action potentials that contain information about stimulus features (intensity, duration, velocity, and location on the body surface). Thus, recording response discharges evoked by defined stimuli in somatosensory neurons allows one to evaluate the biologic significance of anesthetic-induced changes of action potential firing. Facial tactile information is conveyed via the brainstem and the ventral posteromedial (VPM) nucleus of the thalamus to the primary somatosensory cortex (Fig. 1a). This pathway uses the excitatory amino acid glutamate as a transmitter (11). By way of the GABAergic inhibitory input from the thalamic reticular nucleus (TRN), somatosensory information is modulated in VPM thalamocortical relay neurons (TCNs) (10,1214).
Previously we have shown in this model that ISO changes the functional characteristics of thalamic somatosensory information transfer (15). The objective of this study was to test whether this ISO effect is mediated by an enhancement of GABAAergic inhibitory mechanisms. We sought, thereby, to determine to what extent the widely demonstrated in vitro effects of ISO on GABAAergic transmission may contribute to its in vivo effects.
The experiments were performed, after approval of the Hamburg University Animal Research Committee, on eight adult Wistar rats (380550 g body weight) by using methods reported previously (15). Briefly, anesthesia was induced by intraperitoneal injection of ketamine (100 mg/kg). The trachea and femoral vein were cannulated for mechanical ventilation of the lungs and (during the recording sessions) the administration of vecuronium bromide (4 mg · kg-1 · h-1) for muscle relaxation, respectively. End-tidal CO2 concentration and body temperature were continuously monitored and kept in normal ranges. Heart rate was monitored via electrocardiogram. Throughout surgical preparation, anesthesia was maintained by ISO at 1.0%1.6% end-tidal concentration (in oxygen); inspired and end-tidal ISO concentrations were monitored continuously (Capnomac; Datex, Helsinki, Finland). The animals were monitored for adequacy of anesthetic depth, which was based on the absence of movement and heart rate reactions. If such signs were present, ISO was increased immediately. During the recording sessions, the rats were allowed to recover periodically from immobilization to ensure the adequacy of anesthetic depth. The animals head was mounted in a stereotaxic holder with blunt ear bars. A lateral 3-mm square craniotomy was performed, and the dura mater was removed. At the end of the experiments, the animals were killed with an overdose of pentobarbital.
The combined recording/iontophoresis electrodes consisted of a tungsten electrode (impedance at 1 kHz, 2 M The recording/iontophoresis electrode was inserted stereotaxically into the VPM in dorsoventral penetrations, and extracellular single-unit activity was recorded. The neurons association to the VPM was determined by their facial low-threshold receptive fields and the well known features of VPM neurons: the somatotopic sequence of neighboring neurons with small receptive fields and the brisk responses to mechanical stimulation of whiskers, fur, and skin (10,12,17). The neuronal activity was amplified, filtered, and displayed on an oscilloscope from which original recordings were photographed. The data were stored on a digital tape recorder, and off-line analysis was performed by means of a window discriminator, an interface, and a computer program (Spike2 software; Cambridge Electronic Design, Cambridge, UK). Low-threshold mechanoreceptive neurons responding tonically to movement of whiskers were selected (Fig. 1b) (15) and initially characterized under 0.6% or 0.8% ISO. Trapezoidally shaped stimuli were applied to single whiskers glued to the probe of a feedback-controlled electromechanical stimulator (Somedic, Horby, Sweden). Peristimulus time histograms (PSTHs) (bin width 1 ms) were calculated from the neuronal responses to 20 consecutive stimuli (duration, 400600 ms) delivered at 2.5-s intervals. From these PSTHs, neuronal activities were assessed as mean discharge frequencies (spikes per second): the response activity per stimulus was measured for the entire stimulus duration (total response), as well as separately for the ramp (movement response) and the plateau parts (plateau response), and was determined as discharge rate above continuing activity. The continuing activity was measured from the 1-s period immediately before stimulus onset. The same experimental sequence was applied to all neurons:
Equilibration periods for changes of ISO concentration of at least 10 min were observed before neuronal activity was recorded. The final recovery data were taken to ensure that the neuronal discharge characteristics were unchanged over time. The effects of increases in inspired ISO concentrations and the administration of BIC on neuronal activities were analyzed with multiple Wilcoxons tests with Bonferroni corrections, and P < 0.05 was considered significant. With the large ISO concentration, the recurrence of discharges during the plateau part of the stimulus with a rate of at least 50% of baseline plateau response activity (at ISO low) was the criterion for reversal of ON responses to tonic responses during BIC administration (ISO high + BIC).
The effects of clinically relevant concentrations of ISO alone and in the presence of the GABAA-receptor antagonist BIC, administered iontophoretically, were investigated in 23 low-threshold mechanoreceptive TCNs of the rats VPM. The tonically responding TCNs discharge during the plateau of a trapezoidally shaped movement of whiskers, encoding the intensity (displacement amplitude) and duration of the stimulus (plateau response). These neurons also discharge during the ramp part of the stimulus, encoding the velocity of the whisker movement (movement response). An example of a typical tonic response is shown in Figure 1b (upper trace). An increase of ISO from 0.8% to 1.8% resulted in a conversion of the tonic response to an ON response in this neuron, i.e., all stimulus-encoding features of the response are absent but the stimulus onset (Fig. 1b, middle trace). Such a conversion was induced in the population of neurons by increasing ISO (by 0.4%) to 1.0%1.8% (ISO high). The local administration of BIC with ISO high markedly increased the response activity of all neurons investigated. The BIC effects were dose dependent, i.e., they increased with the current and duration of iontophoretic administration. The ON responses were reversed again to tonic responses in 70% of the converted neurons by BIC iontophoresis. Although no reversal to tonic response pattern was seen in the remaining six neurons, BIC effected a marked increase in movement discharge. The original spike records of Figure 2 demonstrate as an example the dose-dependent reversal of the tonic response of a TCN under large ISO concentration with time of BIC administration. The increase in ISO resulted in a conversion of the tonic response present under baseline anesthesia (Fig. 2a) to an ON spike (Fig. 2b). After 7 min of ejection of 60 nA BIC, spikes were elicited during the entire rising ramp of the stimulus, i.e., the forward movement of the whisker (Fig. 2c). Then, spikes also reappeared during the plateau part of the stimulus, and after 15 min of BIC ejection, the tonic response was reestablished with characteristics equal to those under baseline anesthesia (Fig. 2d). After termination of BIC ejection, the suppressant effect of the large ISO dose dominated, and again only ON spikes were elicited (Fig. 2e). The tonic response recovered when the ISO concentration was returned to baseline level (Fig. 2f).
The effects of experimental enhancement and blockade of GABAA receptor function are shown in Figure 3, in which 20 consecutive responses of another TCN were summed for each PSTH. Iontophoretic administration of the GABAA receptor agonist MUSC under baseline anesthesia led to a conversion of the tonic response to a few ON spikes (Fig. 3, a and b), thereby mimicking the effects of an increase of ISO (Fig. 3d). The MUSC-induced inhibition of responses was antagonized by BIC (Fig. 3c). Correspondingly, BIC reversed the ISO-induced suppression of responses (Fig. 3, e and g). This BIC-induced reversal was antagonized by MUSC (Fig. 3f), indicating the involvement of GABAA receptor mechanisms. The recoveries from drug administration and from the large concentration of ISO are demonstrated as well (Fig. 3, h and i).
The quantitative analysis of the responses of all 23 neurons shows the significant suppressive effect of large concentrations of ISO (1.44% ± 0.04%, mean ± SEM) on the total response activity measured from the entire stimulus duration (Fig. 4a). The ISO concentration necessary to induce this suppression was almost double the baseline concentration (0.79% ± 0.01%). The ISO-induced suppression of total response activity (ISO high: mean, 18% of baseline activity) is reversed by BIC (ISO high + BIC: 96%). The same significant suppressive effects of ISO and response reversal by BIC are evident when the different response components in relation to the ramp (29% and 134%, respectively) and plateau parts (5% and 121%, respectively) of the stimulus are analyzed separately (Fig. 4, b and c). The data recorded after termination of BIC administration during ISO high in Figure 4 (ac) show complete recovery from GABAA receptor blockade. The final recovery data were taken after a return to baseline anesthesia at a mean of 2.5 h after the initial baseline recordings of each neuron and demonstrate that the total response magnitude (117%), as well as the movement (123%) and plateau (114%) response components, were unchanged over time.
The increase in ISO concentration almost completely suppressed the continuing activity present between mechanical stimuli (median 5.9 spikes per second [range, 0.221.0 spikes per second] at ISO low, baseline; median 0.0 [range, 0.05.2] spikes per second at ISO high). In contrast to the response activity, the continuing activity did not change significantly during BIC administration (median, 0.2 [range, 0.015.2] spikes per second at ISO high + BIC). This indicates that BIC was administered in doses small enough to avoid extrasynaptically mediated excitation of the neurons.
The main result of our study is that the local thalamic administration of the GABAA receptor antagonist BIC reversed the ISO-induced effects on functional response characteristics in more than two-thirds of TCNs studied. This suggests that the potentiation of GABAAergic inhibitionwhich in vitro studies have demonstrated to be one of ISOs predominant mechanismsmay in fact play an important role in its in vivo anesthetic action. This is the first report that shows that a prototype volatile anesthetic interacts with thalamic inhibitory mechanisms leading to a blockade of sensory information transfer to the cerebral cortex. This block affected specific components of response characteristics, reflecting the ramp and plateau parts of the mechanical stimulus. In deep anesthesia, thereby, the ability of thalamic, and hence also cortical, neurons to discriminate stimulus featuressuch as movement, intensity, and durationdisappears. Information about painful stimuli transmitted by nociceptive neurons may also be blocked by large ISO concentrations at the thalamic level, as preliminary data show (18). The dose-dependent suppressive effects of ISO on TCN continuing and response activity have been demonstrated previously by using systematic 0.2% stepwise increases of ISO concentration (15). Apart from the quantitative reduction in response activity, we also noted an alteration of the response characteristics. Therefore, in this study, we sought to study the mechanisms underlying this effect on the functional response characteristics. Hence, the increase in ISO concentration was targeted this time at that necessary for response conversion. Again, a clear dose-dependent effect was seen because response conversion occurred after almost doubling the baseline ISO concentration. Similarly, doses of iontophoretically administered drugs were targeted at a constant effect on response activity without affecting continuing activity. Drug doses during iontophoresis are difficult to determine because they depend on the time and current of application, the distance and geometry of the pipette tips with respect to the recorded neuron, and the diffusion within the tissue, as well as the uptake mechanisms present; thus, the dose had to be adapted to each neuron (19). When BIC is administered in larger doses, apart from its actions at GABAA receptors, it may directly excite neurons via extrasynaptic actions and thus induce spontaneous (continuing) activity, producing uncontrolled effects. Ketamine was used to induce general anesthesia and to facilitate tracheotomy. After we placed the tracheal catheter, ISO was administered as the only anesthetic for the remainder of the experiment. We consider it unlikely that any residual ketamine effects were still present during neuronal data sampling because the recording sessions commenced at least three hours after the administration of the short-acting induction drug. Additionally, residual ketamine effects would have been noted by comparison of the initial baseline recordings with the recovery recordings (up to four hours later); this comparison, however, revealed no differences (Fig. 4). The GABAergic synapses on the rats TCNs derive exclusively from the TRN (Fig. 1a), because interneurons within VPM are virtually lacking (10,12). Several studies have indicated that inhibitory mechanisms effected by GABAergic axon terminals of TRN neurons control the functional response patterns of VPM neurons; thus, they control the nature of information that is conveyed to the cortex (13). In contrast to the ISO-induced response conversion from tonic to ON responses seen here and in the previous study (15) for TCNs, the tonic response patterns of their afferent inputs (the trigeminothalamic fibers) did not change under ISO concentrations up to 2.0%: only their response magnitude was slightly reduced (15). These results suggest that ISO attenuates the output of somatosensory information of the thalamus, whereas its input is affected to a lesser degree. This is also corroborated by the present findings showing that blockade of thalamic GABAA receptors under large concentrations of ISO did not induce an increase in continuing activity but revealed a recovery of stimulus-evoked response pattern and magnitude in most neurons. Similar findings have been reported from studies of the extracranial somatosensory system (20,21). However, it is not surprising that the thalamic input is affected to some degree by ISO, because others have described effects of volatile anesthetics on dorsal horn neuronal activity (22) that have been ascribed partly to interactions with GABAAergic transmission (23). Therefore, actions of ISO at subthalamic levels may have contributed to the failure to reestablish the tonic response pattern in less than one-third of the TCNs studied, despite thalamic GABAA antagonism. Another potential cause for the preserved ON response pattern despite BIC administration in some TCNs may be a suppression of the glutamatergic signal transmission, because in vitro studies have demonstrated depressive actions of volatile anesthetics on glutamatergic transmission at both presynaptic and postsynaptic sites (2426). However, signal transmission from trigeminothalamic fibers to TCNs expresses a high safety factor under various anesthetic conditions (12,13,27). Another glutamatergic influence onto thalamic neurons is exerted by descending corticothalamic fibers (10,12). This, however, influences both TCNs and TRN neurons and therefore might serve to increase or decrease thalamocortical signal transmission. It is also conceivable that additional remote effects of ISO on the cortical or the brainstem inputs to TRN may enhance its activity. Nonreceptor mediated effects, however, may play a role as well; Ries and Puil (5,6) showed in brain slices that ISO induced a hyperpolarization and shunting of voltage-dependent Na+ and Ca+ currents. Thereby, neuronal excitability was reduced and tonic firing to depolarizing pulses prevented. With respect to our results, it seems, however, that this effect may play only a minor role, because removal of GABAA-receptor-mediated hyperpolarization under a large ISO concentration allowed the neurons to discharge tonically in response to their afferent inputs no differently than under baseline anesthesia. Also, the fact that the continuing activity, definitely suppressed by ISO, was resistant to BIC in doses that reversed the response activity points toward GABAA-receptor-mediated effects. The continuing activity of TCNs, in contrast to their response activity, is independent of GABAergic control from the TRN, because lesioning the TRN had no effect on continuing activity (28). Nevertheless, there is controversy regarding the importance of potential actions of volatile anesthetics on membrane excitability (1,26). There is ample evidence from many in vitro studies showing that enhancement of GABAA-receptor-mediated inhibition may be a major mechanism of action of volatile anesthetics (14). In one study, by using an organotypic neocortical brain slice preparation, Antkowiak (8) demonstrated an inhibition of spontaneous action potential discharges by ISO, which was reversed by more than 90% after the administration of BIC to the bathing solution. In agreement with previous findings from various brain tissue preparations, such as hippocampal brain slices (7) and autaptic cultures of hippocampal neurons (9), these results indicated that the volatile anesthetic acted predominantly on GABAA receptors. Therefore, we conclude from our results that the potentiation of GABAA-receptor-mediated synaptic inhibition seems to play an important role in ISO-induced suppression of TCN activity, outweighing potential effects on other sites, such as glutamatergic synaptic transmission or membrane properties.
Supported in part by the Institute of Physiology and Department of Anesthesiology. We would like to thank Ms. Maren Kurschat for expert technical assistance.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|