Inspiratory premotor neurons in the caudal ventral medulla relay excitatory drive to phrenic and inspiratory intercostal motoneurons in the spinal cord. These neurons are subject to tonic gamma-aminobutyric acid type A (GABA(A))-mediated (GABA(A)ergic) inhibition. In a previous study, 1 minimum alveolar concentration (MAC) halothane depressed overall glutamatergic excitatory drive but did not change overall inhibitory drive to the neurons. This study investigated in further detail the effects of halothane on GABA(A)ergic inhibition by examining postsynaptic GABA(A) receptor activity in these neurons.


Studies were performed in decerebrate, vagotomized, paralyzed, and mechanically ventilated dogs during hypercapnic hyperoxia. The effect of 1 MAC halothane on extracellularly recorded neuronal activity was measured during localized picoejection of the GABA(A) receptor antagonist bicuculline and the GABA(A) agonist muscimol. Complete blockade of GABAergic inhibition by bicuculline allowed estimation of the prevailing overall inhibition of the neuron. The neuronal response to muscimol was used to assess the anesthetic effect on the postsynaptic GABA(A) receptor function.


One minimum alveolar concentration halothane depressed the spontaneous activity of 19 inspiratory premotor neurons by 22.9 +/- 29.1% (mean +/- SD; P < 0.01). Overall excitatory drive was depressed 23.6 +/- 16.9% (P < 0.001). Overall GABAergic inhibition was not changed (+8.7 +/- 27.5%; P = 0.295), but the postsynaptic GABA(A) receptor function was increased by 110.3 +/- 97.5% (P < 0.001).


One minimum alveolar concentration halothane greatly enhanced GABA(A) receptor function on inspiratory premotor neurons but did not change overall synaptic inhibition, indicating that the presynaptic inhibitory input was reduced. Therefore, the anesthetic depression of spontaneous inspiratory premotor neuronal activity in the intact brainstem respiratory network is mainly due to a decrease in overall glutamatergic excitation.

THE caudal ventral respiratory group contains inspiratory and expiratory premotor neurons that provide the excitatory drive to respiratory motoneurons in the spinal cord (fig. 1). The peak level of inspiratory neuronal activity correlates with the respiratory tidal volume. 1Previous studies have shown that 1 minimum alveolar concentration (MAC) halothane reduces the activity of expiratory  premotor neurons by a depression of overall glutamatergic excitatory drive and an increase in overall γ-aminobutyric acid type A (GABAA)–mediated (GABAAergic) inhibition. 2,3The activity of postsynaptic N -methyl-d-aspartate (NMDA) receptors, which are the only physiologically active glutamate receptor subtype in expiratory premotor neurons, 4was not significantly changed. 2The postsynaptic GABAAreceptor function was much more enhanced by the anesthetic than the overall inhibition, indicating that halothane also depressed the presynaptic inhibitory input. 3 

In contrast, a recent study of inspiratory  premotor neurons showed that 1 MAC halothane caused a depression of overall glutamatergic excitation by a depression of the postsynaptic α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) and NMDA receptor responses. 5Presynaptic excitatory drive was not significantly changed. In addition, 1 MAC halothane did not significantly change overall inhibition in these inspiratory neurons.

The current study was performed to clarify whether halothane affected GABAAreceptors in inspiratory neurons. Complete block of GABAergic inhibition with the antagonist bicuculline allowed estimation of the magnitude of prevailing overall GABAergic inhibition of the neuron without anesthesia and at 1 MAC halothane. The exogenous application of the GABAAagonist muscimol onto the same neurons allowed assessment of the GABAAreceptor response at 0 and 1 MAC.

Materials and Methods

Animal Preparation and General Methodology

This research was approved by the Medical College of Wisconsin Animal Care Committee (Milwaukee, Wisconsin) and conformed to standards set forth in the National Institutes of Health Guide for Care and Use of Laboratory Animals. Dogs were induced by mask with isoflurane and intubated with a cuffed endotracheal tube and from then on mechanically ventilated with oxygen. Isoflurane (1.3–1.8 MAC) was applied throughout the surgical procedures and only discontinued after completion of decerebration (1 MAC isoflurane in dogs = 1.4%6). The animals were positioned in a stereotactic device (model 1530; David Kopf Instruments, Tujunga, CA) with the head ventrally flexed (30°). Bilateral neck dissections were performed. The C5 phrenic nerve rootlet was desheathed for recording, and bilateral vagotomy was performed to achieve peripheral deafferentation. This avoids interference of the mechanical ventilation with the underlying central respiratory rhythm and respiratory neuronal activity. Bilateral pneumothorax was performed to minimize brainstem movement and phasic inputs from chest wall mechanoreceptors. The animals were decerebrated at the midcollicular level 7and only then paralyzed (0.1 mg/kg pancuronium, followed by 0.1 mg · kg−1· h−1). An occipital craniotomy was performed to expose the dorsal surface of the medulla oblongata for single neuron recording. Esophageal temperature was maintained at 38.5 ± 1°C. Mean arterial pressure was kept above 100 mmHg and did not differ more than 20% between 0 and 1 MAC. Blood pressure was supported as needed with infusions of phenylephrine (0.5–5 μg · kg−1· min−1). Protocols were only performed during steady state conditions for blood pressure.

Neuron Recording Technique, Data Collection, and Experimental Conditions

Multibarrel compound glass micropipettes consisting of a recording barrel containing a 7-μm carbon filament and three drug barrels were used to simultaneously record extracellular neuronal action potential activity before and during pressure ejection of the GABAAreceptor agonist and antagonist onto inspiratory neurons of the caudal ventral respiratory group. We used the selective GABAAreceptor agonist muscimol (7.5 μm; Research Biochemicals, Natick, MA) and the GABAAantagonist bicuculline methochloride (200 μm; Research Biochemicals), which were dissolved in an artificial cerebrospinal fluid. 2Meniscus changes in the drug barrels were measured to determine the ejected dose rates (resolution, 2 nl). The neurons were located approximately 1.5–3 mm caudal from the obex and 2.5–4.5 mm lateral from the midline. A previous study had shown that approximately 90% of these caudal ventral respiratory group inspiratory neurons were bulbospinal, i.e. , their soma was located in the brainstem and their axons projected to motoneurons in the spinal cord as confirmed with antidromic stimulation techniques. 8Single-cell inspiratory neuronal activity, phrenic nerve activity, picoejection marker pulses, airway carbon dioxide and volatile anesthetic concentrations, systemic blood pressure, and airway pressure were recorded on a digital tape system (model 3000A; A.R. Vetter Co., Rebersburg, PA). These variables or their time averages were also continuously displayed on a computerized chart recorder (Powerlab/16SP; ADInstruments, Castle Hill, Australia). Timing pulses at the beginning and end of neural inspiration were derived from the phrenic neurogram and were used to determine the respiratory phases. The tape-recorded data were digitized and analyzed off-line. Cycle-triggered histograms, triggered at the onset of phrenic activity, were used to quantify the neuronal discharge frequency data.

The protocols were performed under hyperoxic (fraction of inspired oxygen > 0.8) and steady state hypercapnic conditions (arterial carbon dioxide tension [Paco2], 55–65 mmHg). The optimal level of Paco2was adjusted from animal to animal so that adequate phasic phrenic activity during the anesthetic state (1 MAC) was ensured. Great care was taken to keep the Paco2tightly controlled within each neuron protocol. One complete neuron protocol consisted of two sets of two separate picoejection runs (run 1: muscimol; run 2: bicuculline). One set was performed in this sequence at 0, and the other set was performed at 1 MAC halothane (0.9%6). The lack of effect of the artificial cerebrospinal fluid vehicle was tested in a separate run. To maximize the yield of complete neuron protocols, we performed the current protocols with the order of the sets (0 and 1 MAC) randomized, which eliminated the need for end controls. 3,9 

Run 1: Effects of Halothane on Postsynaptic GABAAReceptors

Cycle-triggered histograms (5–20 cycles) were used to obtain values of the average peak neuronal discharge frequency (Fn) for the control period (Fcon) and at each dose rate. The GABAAagonist muscimol was applied in increasing dose rates until a decrease in peak Fnof at least 25 Hz was achieved. Typically, picoejection durations of 6–8 min with 2–3 dose rates were needed.

Run 2: Effects of Halothane on Overall Synaptic Neurotransmission

After recovery from muscimol, the GABAAreceptor antagonist bicuculline was picoejected until complete block of GABAAergic inhibition occurred, i.e. , when an increase in picoejection dose rate did not result in any further increase in peak Fn. Typically, picoejection durations of 5–10 min with several increasing dose rates were required. After the bicuculline run, complete postejection recovery was awaited, which required 30–45 min. Then, the randomized state of anesthesia was changed over approximately 30 min, and after a minimum equilibration time of 15 min, both steps of the protocol were repeated in the same fashion. State of anesthesia refers to either 1 MAC halothane anesthesia or absence of anesthesia (0 MAC).

Statistical Analysis

The data were analyzed for bicuculline first and then for muscimol as follows. During complete GABAAergic block with bicuculline, Fnequals the overall excitatory drive to the neuron (Fe). Under control conditions, the prevailing GABAAergic inhibition reduces Feto Fconby the inhibitory factor α, where α= (Fe− Fcon)/Fe. To calculate the change in overall excitatory drive, the data were normalized to Feat 0 MAC, which was assigned a value of 100%. A two-way repeated-measures analysis of variance was used, with main factors of anesthetic state (0 or 1 MAC) and neurotransmitter status (preejection control vs.  maximal bicuculline block) (SuperANOVA; Abacus Concepts, Inc., Berkeley, CA). The values for Feand Fconwere obtained for the 0 MAC level (Fcon0, Fe0) and the 1 MAC level (Fcon1, Fe1) from the experimental runs. They were then used in the calculation of the anesthetic effect on overall excitation ΔFe(where ΔFe=[Fe1− Fe0]/Fe0) and overall inhibition Δα (where Δα=[α1−α0]/α0).

The effect of 1 MAC halothane on the postsynaptic GABAAreceptor response was quantified by linear regression of Fnon dose rate because previous studies in expiratory neurons have shown the dose–response data for muscimol to be linear 9in the nonsaturating dose rates used in our protocol. This was confirmed for inspiratory neurons. In this regression analysis, the y-intercept was constrained to pass through the Fconvalue at the zero dose rate. Thus, any change in the slope of the regression line reflected the anesthetic-induced change in the dose–response relation. We then normalized the slope to the peak Fnobtained at complete block of GABAergic inhibition (Fe) and designated this normalized slope as the receptor response, ρ. To compare the GABAAreceptor activity at 0 and 1 MAC, the difference of the receptor response relative to the response at 0 MAC was determined for each neuron (i.e. , Δρ=[ρ1−ρ0]/ρ0). Because the data were not normally distributed, a Wilcoxon signed rank test was performed to test whether Δρ was significantly different from no change (StatView; SAS Institute, Inc., Cary, NC).

Presynaptic inhibitory input cannot be measured with our experimental setup, but paired data on overall inhibition (α) and postsynaptic receptor response (ρ) allows estimation of the anesthetic effect on the presynaptic component of inhibition. In short, we assumed that overall inhibition was proportional to the product of presynaptic inhibitory input and postsynaptic inhibitory receptor function (α∝ pre × post). We further assumed that the anesthetic effect on the receptor response ρ was an index for the effect on the endogenously activated postsynaptic receptor function. Therefore, the effect of anesthesia could be expressed as α10= (pre1/pre0) × (ρ10). With Δpre = (pre1− pre0)/pre0, the anesthetic effect on the presynaptic inhibitory input could be calculated as Δpre =[(α10)/(ρ10) − 1] (for details, see  Appendix 1). All results are given as mean ± SD, and P < 0.05 was used to indicate significant differences unless stated otherwise.


Experiments were performed in 22 animals and yielded 19 complete protocols.

Effects of Halothane on Overall Synaptic Neurotransmission

Figure 2shows a representative example of an inspiratory neuronal response to increasing doses of bicuculline at 0 and 1 MAC halothane, respectively. At 0 MAC, bicuculline increased peak Fnfrom 119 to 376 Hz (fig. 3, top ). This means that under control conditions, tonic GABAergic inhibition attenuated the neuronal frequency to 32% of the overall excitatory drive, i.e. , the overall inhibitory factor α0was 0.68. At 1 MAC halothane, bicuculline increased peak Fnfrom 92 to 237 Hz, yielding an α1value of 0.61. Therefore, 1 MAC halothane decreased overall inhibition in this neuron by 10%. The overall excitatory drive to the neuron, Fe, was reduced from 376 Hz to 237 Hz, i.e. , by 37%.

The pooled data for 19 neurons are shown in figure 3, bottom . One minimum alveolar concentration halothane reduced Fconby 22.9 ± 29.1% (P < 0.01) and Feby 23.6 ± 16.9% (P < 0.001). Overall inhibition α was not significantly changed (+8.7 ± 27.5%; P = 0.295).

Effects of Halothane on Postsynaptic GABAAReceptors

The responses of the same inspiratory neuron to muscimol at 0 and 1 MAC halothane are shown in figure 4. In this example, the highest dose rates for muscimol were 0.037 pmol/min at 0 MAC and 0.023 pmol/min at 1 MAC. One minimum alveolar concentration halothane increased the magnitude of the slope of the linear regression line fitted through the dose–response plots from −1,726 to −2,330 Hz · pm−1· min−1(fig. 5). The slopes were normalized to Feat 0 and 1 MAC, i.e. , 376 and 237 Hz, which yielded a receptor response ρ of −460%· pm−1· min−1at 0 MAC and −983%· pm−1· min−1at 1 MAC, i.e. , ρ was increased by 113.7%.

The pooled normalized data from 19 neurons show that 1 MAC halothane significantly enhanced the GABAAreceptor response by 110.3 ± 97.5% (P < 0.001). Estimation of the anesthetic effect on presynaptic inhibitory drive yielded that halothane significantly depressed the presynaptic inhibitory input by 30.8 ± 47.2%. Comparison with the results of a previous study 3showed that the magnitude of the anesthetic effects on the parameters of excitatory and inhibitory neurotransmission did not differ significantly between inspiratory and expiratory premotor neurons (Wilcoxon signed rank test;fig. 6).


The current study shows that halothane greatly enhanced the postsynaptic GABAAreceptor response without altering the overall inhibition of caudal ventral respiratory group inspiratory neurons. This suggests that the anesthetic also reduced the presynaptic inhibitory input (fig. 7). The current study confirms recent findings that the depression of inspiratory premotor neuronal activity by 1 MAC halothane is due to a reduction in overall excitation, whereas overall inhibition does not change. 5 

Halothane Potently Enhances the GABAAReceptor Response

In the current study, 1 MAC halothane greatly enhanced the response of the GABAAreceptor to the agonist muscimol. Like other volatile 10–15and many intravenous anesthetics, 16halothane has been shown to enhance GABAAreceptor function in vitro  by prolongation of the slow component of the inhibitory postsynaptic current. This leads to an increase in negative charge transfer and a depression of postsynaptic neuronal firing. A strong enhancement of the GABAAreceptor response was also observed in expiratory premotor neurons in our decerebrate dog model. 3 

Halothane Depresses the Presynaptic Inhibitory Input

Assuming that overall synaptic inhibition is the product of presynaptic transmitter release and postsynaptic receptor function, we conclude that in this study, the presynaptic inhibitory input was significantly depressed by the anesthetic. This event could also be observed in in vitro  studies, e.g. , in rat hippocampal slices in which halothane depressed the activity of inhibitory interneurons and CA1 pyramidal neurons. 17Depression of inhibitory interneuron discharge activity results in decreased inhibitory synaptic input to the downstream pyramidal neuron. 17Reduction in presynaptic input may result from a decrease in drive to the presynaptic neuron, from depression of axonal conduction, 18or from depression of the presynaptic neurotransmitter release. 19–21 

It is very likely that the reduction in presynaptic inhibitory input in our preparation is not dependent on a single mechanism but represents the product of different effects at several levels of the respiratory network (fig. 1). Our methodology does not allow any conclusion as to the mechanisms of the observed presynaptic anesthetic effect. However, previous studies on expiratory  premotor neurons suggest that this reduction increases with increasing anesthetic concentrations. In decerebrate dogs, the overall inhibition is increased between 10 and 20% by an increase in halothane and sevoflurane from 0 to 1 MAC. 2,3,9GABAAreceptor function was increased by 70–110%, indicating a decrease in presynaptic inhibitory input. 3An increase in halothane from 1 to 2 MAC in neuraxis-intact dogs resulted in a decrease in overall inhibition of 34%. 22Banks and Pearce showed in rat hippocampal slices that the anesthetic-induced increase in GABAAreceptor function was maximal between 1 and 2 MAC for isoflurane and enflurane. 10We hypothesize that in our in vivo  preparation, the enhancement of postsynaptic GABAAreceptor function similarly reached a maximum between 1 and 2 MAC while presynaptic inhibitory drive continuously decreased. Compared to the effects on expiratory neurons, 1 MAC halothane decreased the presynaptic inhibitory input relatively more to inspiratory neurons so that overall synaptic inhibitory input to these neurons was not changed by 1 MAC of the anesthetic.

Role of TASK Channels in the Respiratory System and on Premotor Neurons

Recent in vitro  studies have identified a family of two-pore-domain potassium (TASK) channels that are voltage-independent, hyperpolarizing leak channels that can be modulated by multiple neurotransmitters, such as serotonin and norepinephrine, by a decrease in pH, and also by volatile anesthetics. 23,24,In situ  hybridization showed a widespread prevalence of the TASK-1 channel in the central nervous system, in the locus ceruleus, and in brainstem and spinal cord motoneurons. 24The channel has not been described in the area of the ventral respiratory group. Picoejection of serotonin or norepinephrine on respiratory premotor neurons in the neuraxis-intact canine preparation in our laboratory was without effect. Similarly, picoejection of acidic or alkalotic solutions onto respiratory premotor neurons had no effect on their discharge frequency (unpublished observation, E. Zuperku, Ph.D., Milwaukee, Wisconsin, July 21, 1994, neuronal response data). This suggests that TASK channels are not present on canine respiratory premotor neurons. However, we hypothesize that TASK channels may be present on phrenic and hypoglossal motoneurons in our preparation. This may explain the significantly greater depression of peak phrenic 9and hypoglossal activity (unpublished observation, A. Stucke, M.D., Milwaukee, Wisconsin, February 28 to June 11, 2003, neural activity data) by halothane and sevoflurane, compared with premotor neurons.

TASK channels have also been demonstrated on neurons in regions associated with central chemoreception, such as the locus ceruleus 24,25and the raphe, 26as well as on cells in the carotid bodies. 27,28Although chemoreception may result from the combined activity of a variety of pH-sensitive channels on one neuron, 25the profound impact of clinical concentrations of volatile anesthetics on the conductivity of TASK channels 29,30suggests that these channels play a major role in the anesthetic reduction of chemodrive. Whole cell recordings in rat hypoglossal motoneurons show that the halothane-induced, hyperpolarizing current increases until greater than 2 MAC. 30The anesthetic effect on TASK channels may therefore be an important factor in the reduction of drive to the respiratory system (fig. 1), i.e. , of presynaptic excitatory and inhibitory input to the premotor neurons in our study.

Linearity of Respiratory Premotor Neuronal Responses to Synaptic Input

For our analysis, we used a model of the neuronal discharge activity (Fn), namely Fn= (1 −α) × Fe, in which the interaction between the excitatory (Fe) and inhibitory (α) inputs is not linear but multiplicative (gain modulation). However, in this gain modulation model, Fnis linearly related and proportional to the excitatory drive, Fe. The accuracy of this model applied to bulbospinal respiratory neurons has been demonstrated for a variety of conditions. 31,32 

In the current study, changes in Fnhave been interpreted as changes in excitatory and inhibitory synaptic transmission. However, to accurately quantify these changes, it is mandatory that the neuronal response to synaptic inputs is linear and that the linearity does not change with 1 MAC anesthesia. For example, the response to a given fixed input could vary with changes in neuronal excitability that may occur as a result of changes in overall membrane conductance. Such changes could be brought about by changes in synaptic inputs Feand α. However, evidence exists that this is not the case for respiratory bulbospinal neurons: Picoejection of glutamatergic agonists at a fixed dose rate induced the same net response at markedly different levels of excitatory synaptic drive Fe, resulting from different levels of partial pressure of carbon dioxide ( Appendix 2). Similar results were obtained at different levels in the overall inhibition (α). For example, bicuculline block of the GABAergic gain modulating inhibition resulted in a large increase in the spontaneous Fn, but the AMPA-induced increase in Fnwas unaltered before and after bicuculline. 33In contrast, changes in intrinsic neuronal properties did alter the magnitude of the AMPA-induced responses. Block of the small-conductance Ca2+-activated K+channel, which is responsible for the afterhyperpolarization and therefore controls the neuronal discharge rate, with apamin increased the net response to AMPA. 33This confirmed that changes in excitability can amplify or attenuate these induced responses.

Changes in membrane conductance could also result from a change in K+channel conductance. There is accumulating evidence that volatile anesthetics within the clinical concentration range can alter the conductance of certain types of K+channels, e.g. , the TASK-1 channel. If such changes in membrane conductance were sufficiently large, the excitability of the neuron could be altered, and this in turn would alter the neuronal response to synaptic inputs. We have discussed above that the presence of TASK channels is unlikely in premotor neurons. We believe that the role of this molecular substrate in producing the changes we observed in bulbospinal respiratory premotor neurons is minimal, if any. One minimum alveolar concentration halothane decreased overall excitation in inspiratory neurons and also depressed the postsynaptic AMPA and NMDA receptor responses. 5This could also be explained by hyperpolarization of the neuron so that a given dose rate of the glutamate agonist elicited only a reduced response. However, in expiratory premotor neurons, an increase from 0 to 1 MAC halothane or sevoflurane had no effect on the dose–response relations produced by picoejected NMDA. 2,9Also, sevoflurane had no effect on the NMDA- or AMPA-induced responses of inspiratory premotor neurons, 34suggesting no change in postsynaptic neuronal excitability. Therefore, it is likely that the effects we observed were due to effects of the volatile anesthetics on the receptors involved in synaptic neurotransmission.

Differential Anesthetic Effects on Inspiratory and Expiratory Neurons

The difference in the anesthetic effect on overall inhibition between inspiratory and expiratory premotor neurons emphasizes that the two neuron groups are distinct entities with different synaptic inputs and receptor responses (fig. 1). Both neuron groups are the last relay stations upstream from their respective motoneurons. However, both groups receive excitatory and inhibitory input from different sources and possibly in different magnitudes. 4,35As for the effects of anesthetics, we found that presynaptic  excitatory drive to expiratory neurons was reduced by 1 MAC halothane, whereas it was not changed in inspiratory neurons. 5The magnitude of the effect of halothane on postsynaptic GABAAreceptors seems similar in inspiratory and expiratory neurons. The differential anesthetic effect on overall inhibition may therefore be due to a greater decrease of the presynaptic inhibitory input of the inspiratory neurons, although this difference did not reach statistical significance.

In Vivo  Model Places In Vitro  Findings into Clinical Perspective

The current study confirms that one must be cautious in the transfer of findings from in vitro  preparations obtained under highly reduced conditions to intact neuronal networks operating under physiologic conditions. We find that in inspiratory neurons, the enhancement of GABAAreceptor function, which has been proposed as a major mechanism of general anesthesia, 36is completely offset by a concomitant decrease of the presynaptic inhibitory input. Compared to the significant decrease in overall excitatory drive, the effect on the GABAAreceptor is only one component in the depression of respiration by 1 MAC halothane.

Similarly, the effect of an anesthetic reduction of presynaptic transmitter release at one single synaptic level within a hierarchical neuronal network has to be put into perspective. Several authors report that volatile anesthetics reduce the release of glutamate 19,20,37or GABA 21,38,39from synaptosomes. However, an opposite effect is reported by Nishikawa and MacIver, who describe that in a hippocampal slice preparation, 1 MAC halothane produced an increase  in the frequency of miniature inhibitory postsynaptic potentials in the pyramidal cells, indicating that halothane increased spontaneous GABA release from inhibitory interneurons. 17In addition, differential anesthetic effects on glutamate versus  GABA release may not be reflected in an in vivo  network. In their synaptosomal preparation, Westphalen and Hemmings found that glutamate release was more depressed by isoflurane than GABA release. 21,39However, in our in vivo  inspiratory neuronal preparation, the estimated presynaptic GABAergic inhibition was depressed by 1 MAC halothane, whereas presynaptic glutamatergic excitatory drive was not changed. 5Assuming that this discrepancy is not due to a difference between halothane and isoflurane, this suggests that global anesthetic effects on the respiratory network may be very different from the effects observed on single cellular components.

Methodologic Considerations

The decerebrate animal model allows estimation of anesthetic effects without confounding baseline anesthesia. This is a great advantage because all commonly used background anesthetics have been shown to alter the function of excitatory and inhibitory receptors. 40Even if the background anesthetic does not seem to change global respiratory parameters per se , it will be difficult to exclude complex, e.g. , synergistic interactions between a study anesthetic and an apparently innocuous background anesthetic.

The baseline neuronal activity (i.e. , discharge frequency and pattern) and the magnitude of overall inhibition of premotor neurons in decerebrate dogs are similar to those of neuraxis-intact dogs, which suggests that neurons in both preparations receive comparable excitatory and inhibitory synaptic inputs. 22Also, many other basic respiratory parameters, such as respiratory rate, tachypneic response to halothane, and the response to respiratory reflex inputs, are preserved after decerebration (for a more detailed discussion, see Stuth et al.  2). However, decerebration seems to change inputs to the respiratory rhythm generator because decerebrate dogs tend to become apneic at 1.5 MAC anesthesia (unpublished observations, A. Stucke, M.D., E. Zuperku, Ph.D., and E. Stuth, M.D., Milwaukee, Wisconsin, August 10 to September 9, 1999), whereas in neuraxis-intact dogs, respiratory rhythm is usually preserved until 2 MAC. This prevents us from obtaining anesthetic concentration–response curves in the decerebrate model at supra-MAC doses, which may be helpful in determining the differential anesthetic effects on presynaptic and postsynaptic neurotransmission more clearly.

The picoejection method allows highly localized drug ejection onto the recorded neuron, which is a prerequisite for an unequivocal interpretation of the data. The limitations of this method have been previously delineated. 3,35,41,42Localized effects are further confirmed by the stereotypic response that each agonist/antagonist produces postsynaptically. For example, on the same neuron, bicuculline produces a response that is characteristic of gain modulation, whereas glutamatergic agonists produce parallel baseline shifts in activity. 31NMDA receptor antagonists block tonic excitation, whereas AMPA receptor antagonists block phasic excitation of the same inspiratory neuron. 35,42Such phenomena are observed consistently in all neurons.

Test picoejections with labeled bicuculline in rats showed that the drug concentration in brain tissue steeply declines with distance from the picoejection electrode. 43A theoretical analysis of the diffusion of a drug from a constant point source shows that the concentration decreases inversely (1/distance) with the distance from the source. 44Despite the decrease of drug concentration from the electrode tip, it seems that the ejected bicuculline volume and concentration in our experiments were sufficient to fully block the tonic GABAergic input to the neuron at the highest dose rates, because no further effects were observed with an additional increase in dose rate. The value for overall inhibition, α, has shown great consistency between studies. A previous study of respiratory premotor neurons in neuraxis-intact dogs under 1 MAC halothane used a fourfold larger ejection volume rate of bicuculline with concentrations between 50 and 250 μm and found an overall inhibitory factor, α, of approximately 0.63, 41which was similar to the results of this study (α= 0.60 ± 0.1). On the other hand, the picoejection method does not allow direct determination of the anesthetic effect on the activity of presynaptic neurons, but we must extrapolate the change in presynaptic activity from the ratio of the postsynaptic to overall change in synaptic transmission. The degree of variation in our data may allow small but real anesthetic effects to go undetected. However, prominent effects such as the decrease in the presynaptic inhibitory input or the enhancement of postsynaptic inhibitory receptor function as in the current study can be identified.

In summary, 1 MAC halothane reduced inspiratory neuronal activity by a depression of excitatory drive, whereas overall inhibition was not changed. The prominent enhancement of postsynaptic GABAAreceptor function was fully offset by a depression of presynaptic inhibitory input.

The authors thank Jack Tomlinson (Biological Laboratory Technician, Zablocki Veterans Administration Medical Center, Milwaukee, Wisconsin) for excellent technical assistance.

Appendix 1: Estimation of the Presynaptic Inhibitory Component

We assumed that the magnitude of overall inhibition (α) is proportional to the product of the cascaded presynaptic (pre) and postsynaptic (post) components of inhibitory neurotransmission:


The ratio of overall inhibition for the two anesthetic levels (α10) is a measure of the anesthetic effect on this mechanism. Therefore,


We further assumed that the anesthetic effect on the muscimol response at Dmax(ρ) is an index of the anesthetic effect on the postsynaptic receptor function, i.e. ,


Therefore, equation 4becomes


and rearranging yields

Anesthesia-induced changes (Δ) were expressed as the ratio of the difference between 1 MAC and 0 MAC values, relative to the 0 MAC value. This means for the presynaptic inhibitory component Δpre:

Substituting equation 7into equation 8yields


which estimates the anesthetic effect on the presynaptic component based on the effects on overall inhibition and on the postsynaptic component of neurotransmission.

Appendix 2: Linearity of Neuronal Responses

In these studies, changes in neuronal discharge frequency have been used to quantify changes in the synaptic inputs and their postsynaptic responses. On the basis of this assumption, the results will be accurate and reliable if the output of the neuron is a linear function of the inputs. By definition, a linear system must demonstrate the following property:


where F is a linear function. This property states that the output to the combined input, x1+ x2, must be equal to the sum of the outputs to each input when applied separately. To simplify notation, let y1= F(x1) and y2= F(x2). Then,

For the specific case relevant to this study of respiratory bulbospinal neurons, let Δ= y2be the net response to the exogenous application of NMDA at a constant dose rate. Then, equation 2becomes

Solving equation 3for Δ yields:

Under the condition of linearity, both sides of equation 4must be equal. Therefore, for the linear case, Δ remains constant at any value of y1. That is, the magnitude of the net response does not change with changes in the output level of the neuron.

The following experimental data support the linearity of the response of respiratory bulbospinal neurons. At different levels of excitatory drive, the plots of peak discharge frequency, Fn, versus  NMDA dose rate are parallel. In addition, because these plots are linear over a dose rate range that produces marked increases in Fn, their slopes are parallel (fig. 8). Therefore, at a given dose rate, the net responses, Δ1and Δ2, are not significantly different from each other, even though marked differences are evident in the initial Fnbaseline levels (fig. 8, arrows ).

Further evidence of response linearity is indicated by the upward parallel shift in the discharge patterns of respiratory bulbospinal neurons. Figure 9shows cycle-triggered histograms of an inspiratory augmenting neuron before (thick line ) and during steady state picoejection of NMDA (thin line ). The vertical dashed lines  delineate the active phase. For the active phase, the plot of time-aligned Fnduring NMDA versus  control Fnis linear with a slope of 1.01 and a y-intercept of 56.5 (fig. 9, right ). This indicates an upward shift of 56.5 Hz and confirms that the shift is parallel. The addition of 56.5 Hz to the control cycle-triggered histograms results in a pattern that precisely coincides with the cycle-triggered histograms of Fnduring NMDA (fig. 9, left , triangles ). These data indicate that the net response to NMDA, Δ, is the same throughout the inspiratory phase, when the output of the neuron increased from 45 to 130 Hz. The NMDA-induced increase in activity during the normally silent expiratory phase is somewhat less than that during the active phase because of the sub–firing threshold level of the neuron during the silent phase.

The above results are consistent with the finding of a linear relation between the level of intracellularly injected DC current and the discharge frequency of medullary respiratory neurons in cats. 45These results suggest that the magnitude of the current, which reaches the soma/spike-initiation zone of medullary respiratory neurons, is linearly encoded into the discharge frequency. In addition, the NMDA results presented above suggest that the synaptic/receptor response is linearly encoded into the somatic current.


Eldridge FL: Relationship between respiratory nerve and muscle activity and muscle force output. J Appl Physiol 1975; 39: 567–74
Stuth EAE, Krolo M, Stucke AG, Tonkovic-Capin M, Tonkovic-Capin V, Hopp FA, Kampine JP, Zuperku EJ: Effects of halothane on excitatory neurotransmission to medullary expiratory neurons in a decerebrate dog model. A nesthesiology 2000; 93: 1474–81
Stucke AG, Stuth EAE, Tonkovic-Capin V, Tonkovic-Capin M, Hopp FA, Kampine JP, Zuperku EJ: Effects of halothane and sevoflurane on inhibitory neurotransmission to medullary expiratory neurons in a decerebrate dog model. A nesthesiology 2002; 96: 955–62
Dogas Z, Stuth EAE, Hopp FA, McCrimmon DR, Zuperku EJ: NMDA receptor-mediated transmission of carotid body chemoreceptor input to expiratory bulbospinal neurones in dogs. J Physiol (Lond) 1995; 487: 639–51
Stucke AG, Zuperku EJ, Tonkovic-Capin V, Tonkovic-Capin M, Hopp FA, Kampine JP, Stuth EAE: Halothane depresses glutamatergic neurotransmission to brainstem inspiratory premotor neurons in a decerebrate dog model. A nesthesiology 2003; 98: 897–905
Kazama T, Ikeda K: Comparison of MAC and the rate of rise of alveolar concentration of sevoflurane with halothane and isoflurane in the dog. A nesthesiology 1988; 68: 435–7
Tonkovic-Capin M, Krolo M, Stuth EAE, Hopp FA, Zuperku EJ: Improved method of canine decerebration. J Appl Physiol 1998; 85: 747–50
Stuth EAE, Tonkovic-Capin M, Kampine JP, Bajic J, Zuperku EJ: Dose-dependent effects of halothane on the carbon dioxide responses of expiratory and inspiratory bulbospinal neurons and the phrenic nerve activities in dogs. A nesthesiology 1994; 81: 1470–83
Stucke AG, Stuth EAE, Tonkovic-Capin V, Tonkovic-Capin M, Hopp FA, Kampine JP, Zuperku EJ: Effects of sevoflurane on excitatory neurotransmission to medullary expiratory neurons and on phrenic nerve activity in a decerebrate dog model. A nesthesiology 2001; 95: 485–91
Banks MI, Pearce RA: Dual actions of volatile anesthetics on GABAAIPSCs: Dissociation of blocking and prolonging effects. A nesthesiology 1999; 90: 120–34
de Sousa SLM, Dickinson R, Lieb WR, Franks NP: Contrasting synaptic actions of the inhalational general anesthetics isoflurane and xenon. A nesthesiology 2000; 92: 1055–66
Jones MV, Harrison NL: Effects of volatile anesthetics on the kinetics of inhibitory postsynaptic currents in cultured rat hippocampal neurons. J Neurophysiol 1993; 70: 1339–49
Lin L-H, Chen LL, Zirrolli JA, Harris RA: General anesthetics potentiate γ-aminobutyric acid actions on γ-aminobutyric acidAreceptors expressed by xenopus oocytes: Lack of involvement of intracellular calcium. J Pharmacol Exp Ther 1992; 263: 569–78
Jones MV, Brooks PA, Harrison N: Enhancement of γ-aminobutyric acid-activated Cl-currents in cultured rat hippocampal neurones by three volatile anaesthetics. J Physiol (Lond) 1992; 449: 279–93
Nishikawa K, MacIver MB: Agent-selective effects of volatile anesthetics on GABAA receptor-mediated synaptic inhibition in hippocampal interneurons. A nesthesiology 2001; 94: 340–7
Pistis M, Belelli D, Peters JA, Lambert JJ: The interaction of general anaesthetics with recombinant GABAA and glycine receptors expressed in Xenopus laevis oocytes: A comparative study. Br J Pharmacol 1997; 122: 1707–19
Nishikawa K, Maclver MB: Membrane and synaptic actions of halothane on rat hippocampal pyramidal neurons and inhibitory interneurons. J Neurosci 2000; 20: 5915–23
Mikulec AA, Pittson S, Amagasu SM, Monroe FA, MacIver MB: Halothane depresses action potential conduction in hippocampal axons. Brain Res 1998; 796: 231–8
Miao N, Frazer MJ, Lynch C III: Volatile anesthetics depress Ca2+transients and glutamate release in isolated cerebral synaptosomes. A nesthesiology 1995; 83: 593–603
Schlame M, Hemmings HC: Inhibition by volatile anesthetics of endogenous glutamate release from synaptosomes by a presynaptic mechanism. A nesthesiology 1995; 82: 1406–16
Westphalen RI, Hemmings HC: Selective depression by general anesthetics of glutamate versus GABA release from isolated cortical nerve terminals. J Pharmacol Exp Ther 2003; 304: 1188–96
Stuth EAE, Krolo M, Tonkovic-Capin M, Hopp FA, Kampine JP, Zuperku EJ: Effects of halothane on synaptic neurotransmission to medullary expiratory neurons in the ventral respiratory group of dogs. A nesthesiology 1999; 91: 804–14
Patel AJ, Honore E, Lesage F, Fink M, Romey G, Lazdunski M: Inhalational anesthetics activate two-pore-domain background K channels. Nat Neurosci 1999; 2: 422–6
Talley EM, Lei Q, Sirois JE, Bayliss DA: TASK-1, a two-pore domain K channel, is modulated by multiple neurotransmitters in motoneurons. Neuron 2000; 25: 399–410
Filosa JA, Putnam RW: Multiple targets of chemosensitive signaling in locus coeruleus neurons: Role of K+ and Ca2+ channels. Am J Physiol Cell Physiol 2003; 284: C145–55
Washburn CP, Sirois JE, Talley EM, Guyenet PG, Bayliss DA: Serotonergic raphe neurons express TASK channel transcripts and a TASK-like pH- and halothane-sensitive K+ conductance. J Neurosci 2002; 22: 1256–65
Buckler KJ, Williams BA, Honore E: An oxygen-, acid- and anaesthetic-sensitive TASK-like background potassium channel in rat arterial chemoreceptor cells. J Physiol 2000; 525.1: 135–42
Yamamoto Y, Kummer W, Atoji Y, Suzuki Y: TASK-1, TASK-2, TASK-3 and TRAAK immunoreactivities in the rat carotid body. Brain Res 2002; 950: 304–7
Bayliss DA, Talley EM, Sirois JE, Lei Q: TASK-1 is a highly modulated pH-sensitive “leak” K+ channel expressed in brainstem respiratory neurons. Respir Physiol 2001; 129: 159–74
Sirois JE, Lie Q, Talley EM, Lynch C III, Bayliss DA: The TASK-1 two-pore domain K+channel is a molecular substrate for neuronal effects of inhalation anesthetics. J Neurosci 2000; 201: 6347–54
McCrimmon DR, Zuperku EJ, Hayashi F, Dogas Z, Hinrichsen CFL, Stuth EA, Tonkovic-Capin M, Krolo M, Hopp FA: Modulation of the synaptic drive to respiratory premotor and motor neurons. Respir Physiol 1997; 110: 161–76
Zuperku EJ, McCrimmon DR: Gain modulation of respiratory neurons. Respir Physiol Neurobiol 2002; 131: 121–33
Tonkovic-Capin V, Stucke AG, Stuth EA, Tonkovic-Capin M, Hopp FA, McCrimmon DR, Zuperku EJ: Differential processing of excitation by GABAergic gain modulation in canine caudal ventral respiratory group neurons. J Neurophysiol 2003; 89: 862–70
Stucke AG, Stuth EAE, Tonkovic-Capin V, Kampine JP, Zuperku EJ: Effects of sevoflurane (SEVO) on excitatory neurotransmission to medullary inspiratory (I) neurons in a decerebrate dog model. FASEB J 2003: 302.2
Krolo M, Stuth EA, Tonkovic-Capin M, Dogas Z, Hopp FA, McCrimmon DR, Zuperku EJ: Differential roles of ionotropic glutamate receptors in canine medullary inspiratory neurons of the ventral respiratory group. J Neurophysiol 1999; 82: 60–8
Antkowiak B: Different actions of general anesthetics on the firing patterns of neocortical neurons mediated by the GABAAreceptor. A nesthesiology 1999; 91: 500–11
Kirson ED, Yaari Y, Perouansky M: Presynaptic and postsynaptic actions of halothane at glutamatergic synapses in the mouse hippocampus. Br J Pharmacol 1998; 124: 1607–14
Hemmings HC, Westphalen RI: Effects of isoflurane and propofol on the release of [3H] glutamate and [14C] GABA from isolated nerve terminals (abstract). A nesthesiology 2001; 95: A–794
Westphalen RI, Hemmings HC: Selective depression by isoflurane and propofol of glutamate vs. GABA release from isolated cortical nerve terminals (abstract). A nesthesiology 2002; 97: A–818
Hara K, Harris RA: The anesthetic mechanism of urethane: the effects on neurotransmitter-gated ion channels. Anesth Analg 2002; 94: 313–8
Dogas Z, Krolo M, Stuth EA, Tonkovic-Capin M, Hopp FA, McCrimmon DR, Zuperku EJ: Differential effects of GABAAreceptor antagonists in the control of respiratory neuronal discharge patterns. J Neurophysiol 1998; 80: 2368–77
Krolo M, Stuth EA, Tonkovic-Capin M, Hopp FA, McCrimmon DR, Zuperku EJ: Relative magnitude of tonic and phasic synaptic excitation of medullary inspiratory neurons in dogs. Am J Physiol Regulatory Integrative Comp Physiol 2000; 279: R639–49
Segura T, Martin DS, Sheridan PJ, Haywood JR: Measurement of the distribution of [3H]bicuculline microinjected into the rat hypothalamus. J Neurosci Methods 1992; 41: 175–86
Stone TW: Microiontophoresis and Pressure Ejection. IBRO Handbook Series. Methods in the Neurosciences. Vol 8. Edited by Smith AD. Chichester, John Wiley and Sons, 1985, pp 39–75
Richter DW, Heyde F: Accommodative reactions of medullary respiratory neurons of the cat. J Neurophysiol 1975; 38: 1172–80
Ezure K: Synaptic connections between medullary respiratory neurons and considerations on the genesis of respiratory rhythm. Prog Neurobiol 1990; 35: 429–50