Sevoflurane Enhances γ-Aminobutyric Acid Type A Receptor Function and Overall Inhibition of Inspiratory Premotor Neurons in a Decerebrate Dog Model

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 the Care and Use of Laboratory Animals .9Anesthesia was induced by mask with isoflurane, and the dogs were intubated with a cuffed endotracheal tube and from then on were 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%).10The 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 level11and 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.

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 GABA^Areceptor agonist and antagonist onto inspiratory neurons of the caudal ventral respiratory group. We used the selective GABA^Areceptor agonist muscimol (7.5 μm; Research Biochemicals, Natick, MA) and the GABA^Aantagonist bicuculline methochloride (200 μm; Research Biochemicals), which were dissolved in an artificial cerebrospinal fluid.6Meniscus 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.12Single-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). A time–amplitude window discriminator was used to produce rate-meter recordings (Fⁿ: average discharge frequency per 100-ms period) of the neuronal activity. 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 (average Fⁿper 50-ms bin), 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 sevoflurane (= 2.4%).10The 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.7,8 

Cycle-triggered histograms (5–20 cycles) were used to obtain values of the average peak neuronal discharge frequency (Fⁿ) for the control period (F^con) and at each dose rate. The GABA^Aagonist 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 two to three dose rates were needed.

After recovery from muscimol, the GABA^Areceptor antagonist bicuculline was picoejected until complete block of GABA^Aergic 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 sevoflurane anesthesia or absence of anesthesia (0 MAC).

The data were analyzed for bicuculline first and then for muscimol as follows. During complete GABA^Aergic block with bicuculline, Fⁿequals the overall excitatory drive to the neuron (F^e).4,7Under control conditions, the prevailing GABA^Aergic inhibition reduces F^eto F^conby the inhibitory factor α, where α= (F^e− F^con)/F^e. To calculate the change in overall excitatory drive, the data were normalized to F^eat 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 F^eand F^conwere obtained for the 0 MAC level (F^con0, F^e0) and the 1 MAC level (F^con1, F^e1) from the experimental runs. They were then used in the calculation of the anesthetic effect on overall excitation ΔF^e(where ΔF^e=[F^e1− F^e0]/F^e0) and overall inhibition Δα (where Δα=[α¹−α⁰]/α⁰).

The effect of 1 MAC sevoflurane on the postsynaptic GABA^Areceptor 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 linear7in the nonsaturating dose rates used in our protocol. This was confirmed for inspiratory neurons.4In this regression analysis, the y-intercept was constrained to pass through the F^convalue at the zero dose rate. Therefore, any change in the slope of the regression line reflected the anesthetic-induced change in the dose–response relation. Using this approach, the amount of reduction in Fⁿdue to muscimol is proportional to the slope and can be determined when the dose rate is specified. Furthermore, when the magnitude of the muscimol-induced reduction of Fⁿis compared at the same dose rate as a ratio, the dose rate term cancels, and the ratio is independent of dose rate. Therefore, the ratio of slopes is equivalent to the ratio of Fⁿreductions at any given dose rate. We then normalized the dose–response slope to the peak Fⁿobtained at complete block of GABAergic inhibition (F^e) for each anesthetic state and designated this normalized slope as the receptor response, ρ. To compare the GABA^Areceptor 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).

The presynaptic inhibitory input cannot be measured with our experimental setup, but paired data from the same neuron for 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 (pre) and postsynaptic inhibitory receptor function (post; then 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 α¹/α⁰= (pre¹/pre⁰)· (ρ¹/ρ⁰). With Δpre = (pre¹− pre⁰)/pre⁰, the anesthetic effect on the presynaptic inhibitory input could be calculated as Δpre =[(α¹/α⁰)/(ρ¹/ρ⁰) − 1] (for details, see  appendix). All results are given as mean ± SD, and P < 0.05 was used to indicate significant differences unless stated otherwise.

Experiments were performed in 23 animals and yielded 21 complete protocols.

Figure 1shows a representative example of an inspiratory neuronal response to increasing doses of bicuculline at 0 and 1 MAC sevoflurane, respectively. At 0 MAC, bicuculline increased peak Fⁿfrom 123 to 258 Hz (fig. 2, top). This means that under control conditions, tonic GABAergic inhibition attenuated the neuronal frequency to 48% of the overall excitatory drive, i.e. , the overall inhibitory factor α⁰was 0.52. At 1 MAC sevoflurane, bicuculline increased peak Fⁿfrom 76 Hz to 240 Hz, yielding an α¹value of 0.68. Therefore, 1 MAC sevoflurane increased overall inhibition in this neuron by 31%. The overall excitatory drive to the neuron, F^e, was reduced from 258 Hz to 240 Hz, i.e. , by 7%. Control frequency, F^con, was decreased by 37.5%.

The pooled data for 21 neurons are shown in figure 2, bottom. One MAC sevoflurane reduced F^conby 32.6 ± 20.5% (P < 0.001) and F^eby 17.9 ± 19.8% (P < 0.01). Overall inhibition α increased from 0.53 ± 0.1 to 0.62 ± 0.1, i.e. , by 18.5 ± 18.2% (P < 0.001).

The response of the same inspiratory neuron to muscimol at 0 and 1 MAC sevoflurane is shown in figure 3. In this example, the highest dose rates for muscimol were 0.092 pmol/min at 0 MAC and 0.024 pmol/min at 1 MAC. One MAC sevoflurane increased the magnitude of the slope of the linear regression line fitted through the muscimol dose–response plots from −817 Hz·pM⁻¹·min⁻¹to −1,270 Hz·pM⁻¹·min⁻¹(fig. 4). The slopes were normalized to F^eat 0 and 1 MAC, i.e. , 258 and 240 Hz, which yielded a normalized receptor response ρ of −3.17·pM⁻¹·min⁻¹at 0 MAC and -5.29 ·pM⁻¹·min⁻¹at 1 MAC, i.e. , ρ was increased by 67%.

One neuron was removed from the calculation of the pooled normalized data because the value for Δρ was more than 2 SDs above the mean. The pooled data from 20 neurons (fig. 2, bottom graph) show that 1 MAC sevoflurane significantly enhanced the GABA^Areceptor response by 184.4 ± 121.8% (P < 0.001). Estimation of the anesthetic effect on presynaptic inhibitory drive indicated that sevoflurane significantly depressed the presynaptic inhibitory input by 49.7 ± 24.1% (P < 0.001; fig. 2, bottom).

The uniform experimental design among studies allowed pooling data from this study with our previous studies3,4,6–8and our companion article in this issue of Anesthesiology;5 table 1). This allows us to highlight (1) differences in neurotransmitter control between inspiratory and expiratory premotor neurons, (2) differential effects of halothane and sevoflurane on inspiratory neurons, and (3) differential effects of sevoflurane on the two different premotor neuron types (Mann–Whitney U test).

In the absence of anesthesia (0 MAC), bicuculline block revealed that the overall inhibitory input to inspiratory neurons was significantly greater (α= 0.57 ± 0.1, n = 84; P < 0.001) than to expiratory neurons (α= 0.47 ± 0.1, n = 65; P < 0.001; fig. 5), which was reflected in higher peak discharge frequencies F^eof inspiratory neurons (268.5 ± 83.2 Hz, n = 84) compared with expiratory neurons (241.8 ± 77.6 Hz, n = 65; P < 0.05; fig. 5). Control frequency (F^con) was not different between the two neuron groups.

For inspiratory neurons, 1 MAC sevoflurane decreased neuronal control frequency significantly more (−31.2 ± 23.6, n = 44) than 1 MAC halothane (−21.7 ± 23.6, n = 40; P < 0.05; table 1and fig. 6). Also, 1 MAC sevoflurane enhanced overall inhibition significantly more (+15.0 ± 22.1, n = 44) than 1 MAC halothane (+6.8 ± 21.6, n = 40; P < 0.05). Both anesthetics depressed overall excitatory drive to a similar extent (fig. 6).

Lastly, 1 MAC sevoflurane enhanced the GABA^Areceptor function in inspiratory neurons (+184.4 ± 121.8%, n = 20) significantly more than in expiratory neurons (+65.0 ± 70.9%, n = 23; P < 0.001; table 1).

The current study confirms that 1 MAC sevoflurane depresses neuronal activity in inspiratory premotor neurons in vivo  by a depression of overall excitation and an enhancement of overall inhibition. In addition, we now show that the increase in overall inhibition results from a marked anesthetic enhancement of GABA^Areceptor function, which allows the conclusion that presynaptic inhibitory input was reduced by the anesthetic (fig. 7).

Our results agree with in vitro  studies that consistently show an enhancement of GABA^Areceptor function by volatile anesthetics. Studies on rat hippocampal slices show that this enhancement is characterized by a reduction of the peak inhibitory postsynaptic current together with a large prolongation of the duration of the inhibitory current, which leads to a net increase in inhibitory charge transfer.13,14A pronounced enhancement of GABA^Areceptor function by halothane and sevoflurane was also found in our previous in vivo  studies on expiratory neurons and in inspiratory neurons with halothane.4,7Interestingly, 1 MAC sevoflurane enhanced GABA^Areceptor function significantly more in inspiratory than expiratory neurons.

We conceptualize that overall inhibition is the product of presynaptic inhibitory input and postsynaptic GABA^Areceptor function in our neuronal paradigm. This suggests that the discrepancy between the pronounced enhancement of postsynaptic GABA^Areceptor function and the much lesser increase in overall inhibition results from a depressant effect of sevoflurane on the presynaptic inhibitory input. In vitro  studies suggest several mechanisms of such presynaptic depression, which may play a role in the in vivo  model. Volatile anesthetics have been shown to depress GABA transmitter release from synaptosomes and may thus reduce inhibitory neurotransmission.15In addition, in a rat hippocampal slice preparation, inhibitory interneurons were depressed by volatile anesthetics.14,16One MAC-equivalent halothane or sevoflurane more than doubled the negative charge transfer to inhibitory interneurons with no difference between the anesthetics.14In the same preparation, in CA1 pyramidal neurons that are downstream from the inhibitory interneurons halothane and sevoflurane caused a decrease in excitatory postsynaptic potential amplitude and an increase in inhibitory postsynaptic potential amplitude,14resulting in an increased failure rate to elicit action potentials.16This suggests that the presynaptic inhibitory input to as well as the activity of postsynaptic CA1 pyramidal neurons could be simultaneously depressed by volatile anesthetics.

The pooled data from our in vivo  preparation show that sevoflurane enhanced overall inhibition significantly more than halothane, whereas the effect on the postsynaptic GABA^Areceptor was not different.4This suggests that halothane had a greater depressant effect on the presynaptic inhibitory input. The depression of overall excitatory drive did not differ between the anesthetics, but neuronal control frequency was depressed more by 1 MAC sevoflurane than by halothane.3–5The differential anesthetic depression of inspiratory premotor neurons may explain why in a previous study 1 MAC sevoflurane depressed peak phrenic activity significantly more than 1 MAC halothane.8The phrenic neurogram represents the magnitude of the inspiratory neural output of the central respiratory network that reaches the phrenic motor neurons, which convey drive to the inspiratory muscles. It can be regarded as a neural index of the magnitude of tidal volume.17A depression of peak phrenic activity reflects the anesthetic depression of respiratory drive that is observed clinically.18Because peak phrenic activity was depressed by 43% with 1 MAC halothane and 60% with 1 MAC sevoflurane, i.e. , significantly more than inspiratory neuronal activity,8it is likely that there is an additional anesthetic depression of neuronal activity at the level of the phrenic motoneurons, which also differs between the two anesthetics.

The in vivo  preparation allows investigation of respiratory premotor neurons as part of a functional respiratory network that generates respiratory drive under close to physiologic conditions. We have discussed the advantages and limitations of the decerebrate preparation in detail in our companion article5and elsewhere.4We have also elaborated the reliability, advantages, and limitations of the picoejection method before.2,3,7With the use of pooled data in this study, we reach a higher number of total protocols, which increases the sensitivity of the analysis and allows distinction between anesthetic effects and physiologic differences of smaller magnitude. All studies on inspiratory and expiratory premotor neurons were conducted under the same experimental conditions. The results, i.e. , anesthetic effects on overall excitation or inhibition, from experiments on the same neuron type did not differ between the separate studies (data not shown) and were therefore pooled.

The synopsis of all studies conducted with halothane and sevoflurane on inspiratory and expiratory premotor neurons emphasizes the dual effect of volatile anesthetics on the respiratory system. Volatile anesthetics reduce glutamatergic excitation as well as presynaptic inhibitory input to respiratory premotor neurons. This depression seems dose dependent because in the neuraxis-intact dog, an increase in halothane from 1 to 2 MAC leads to an additional depression of overall excitation by 35% and a depression of overall inhibition of 34%.19In contrast, we hypothesize that the pronounced enhancement of GABA^Areceptor function that we consistently see in both neuron types and with both anesthetics may reach a maximum around 1 MAC. This hypothesis is supported by in vitro  observations in a rat hippocampal slice preparation.13Overall, only in vivo  preparations allow the study of anesthetic effects on neurons in an environment where they are exposed to the complex, physiologic tonic and phasic synaptic inputs. Our data provide an essential link between the mechanistic study of anesthetics on single receptors or single neurons in isolation in vitro  and the purely descriptive study of the respiratory side effects of volatile anesthetics that are observed in clinical practice.

In summary, 1 MAC sevoflurane increased overall inhibition to inspiratory premotor neurons by a pronounced enhancement of the postsynaptic GABA^Areceptor function, whereas presynaptic inhibitory input was decreased.

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

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 (α¹/α⁰) is a measure of the anesthetic effect on this mechanism. Thus,

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

Thus, equation 3bbecomes

and rearranging yields

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

Substituting equation 6into equation 7yields

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