Cholinergic Modulation of Sevoflurane Potency in Cortical and Spinal

Home , Bicuculline, Strychnine

Anesthesiology 2007; 106:1147–55 Copyright © 2007, the American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc. Cholinergic Modulation of Sevoﬂurane Potency in Cortical and Spinal Networks In Vitro Christian Grasshoff, M.D.,* Berthold Drexler, M.D.,* Harald Hentschke, Ph.D.,† Horst Thiermann, M.D.,‡ Bernd Antkowiak, Ph.D.§

Background: Victims of organophosphate intoxication with dred thousand casualties every year.6 Scenarios of mass injury cholinergic crisis may have need for sedation and anesthesia, such as the terrorist attack in Tokyo in 1995 with the nerve but little is known about how anesthetics work in these pa- agent sarin occur rarely. However, when they become reality, tients. Recent studies suggest that cholinergic stimulation impairs ␥-aminobutyric acid type A (GABA ) receptor function. victims are likely to suffer not only from intoxication but also A 1,7 Because GABAA receptors are major targets of general anesthet- from physical trauma. These subjects may have to undergo Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 ics, the authors investigated interactions between acetylcholine surgical interventions, raising the need for general anesthesia.8 and sevoflurane in spinal and cortical networks. So far, little is known about how general anesthetics act Methods: Cultured spinal and cortical tissue slices were ob- in patients afflicted with cholinergic crisis. This problem is tained from embryonic and newborn mice. Drug effects were assessed by extracellular voltage recordings of spontaneous of relevance because there is clear evidence in the litera- ␥ action potential activity. ture that -aminobutyric acid type A (GABAA) receptor– Results: Sevoflurane caused a concentration-dependent de- mediated inhibition, a molecular mechanism that is in- ؍ crease in spontaneous action potential firing in spinal (EC50 volved in mediating the sedative and hypnotic actions of .mM) slices 0.01 ؎ 0.29 ؍ mM) and cortical (EC 0.02 ؎ 0.17 50 most clinically used anesthetics, is hampered by cholin- Acetylcholine elevated neuronal excitation in both preparations and diminished the potency of sevoflurane in reducing ergic stimulation: In rat cerebral cortical slices, it was dem- action potential firing in cortical but not in spinal slices. This onstrated that acetylcholine reduces ␥-aminobutyric acid brain region-specific decrease in sevoflurane potency was mim- (GABA) release from presynaptic terminals by activating 9 icked by the specific GABAA receptor antagonist bicuculline, muscarinic receptors. Furthermore, a recent study has suggesting that (1) GABA receptors are major molecular targets A shown that GABA release is also decreased via nicotinic for sevoflurane in the cortex but not in the spinal cord and (2) receptors.10 Besides these presynaptic actions of acetylcho- acetylcholine impairs the efficacy of GABAA receptor–mediated inhibition. The latter hypothesis was supported by the finding line, further mechanisms may come into play, rendering that acetylcholine reduced the potency of etomidate in depress- GABAA receptor–mediated inhibition ineffective. Excessive ing cortical and spinal neurons. neuronal activity, as observed during cholinergic crisis, Conclusions: The authors raise the question whether cholin- shifts the equilibrium potential for chloride ions toward ergic overstimulation decreases the efficacy of GABAA receptor function in patients with organophosphate intoxication, more positive values, thereby reducing the amplitude of 11,12 thereby compromising anesthetic effects that are mediated pre- GABAA receptor–mediated synaptic events. In addi- dominantly via these receptors such as sedation and hypnosis. tion, extracellular accumulation of GABA may desensitize

synaptically located GABAA receptors and, as a conse- MANY potent insecticides and nerve agents belong to the quence, depress synaptic GABAergic transmission. family of organophosphorus compounds. These highly toxic We have previously shown that volatile anesthetics chemicals act by blocking acetylcholinesterase activity, such as halothane, isoflurane, and enflurane depress neu- thereby causing a potentially life-threatening cholinergic crisis ronal activity in cortical networks in vivo and in vitro by hallmarked by centrally mediated symptoms such as general- potentiating GABAA receptor–mediated synaptic inhibi- ized convulsions, respiratory failure, and cardiovascular insta- tion at concentrations causing sedation and hypno- bility.1–3 Furthermore, peripheral symptoms including rhinor- sis.13,14 Assuming that actions on the molecular and rhea, hypersalivation, bronchoconstriction, and network level are linked causally, how does a cholin- 1,4,5 neuromuscular block are commonly observed. Suicidal ergic-induced suppression of GABAA receptor function and accidental poisoning by organophosphates is a wide- affect the potency of general anesthetics in depressing spread problem in the developing world, causing several hun- neuronal activity? In the current study, interactions between acetylcholine and sevoflurane were analyzed be- * Research Assistant, † Assistant Professor, § Professor of Experimental Anes- cause volatile anesthetics were recommended for main- thesiology, Experimental Anesthesiology Section, Department of Anesthesiology, tenance of anesthesia in nerve agent–intoxicated University of Tuebingen, Tuebingen, Germany. ‡ Postdoctoral University Lec- 1 turer of Pharmacology and Toxicology, GAF Institute of Pharmacology and patients. Experiments were conducted in organotypic Toxicology, Munich, Germany. slice cultures derived from the neocortex and spinal Received from the Experimental Anesthesiology Section, Department of An- cord because these neuronal microcircuits are important esthesiology and Intensive Care, Eberhard-Karls-University, Tuebingen, Germany. Submitted for publication September 25, 2006. Accepted for publication Febru- substrates for producing major components of general ary 9, 2007. Supported in part by contract No. M SAB1 3A014 from the German anesthesia such as amnesia, hypnosis, and immobili- Ministry of Defense, Munich, Germany. 14–20 Address correspondence to Dr. Grasshoff: Experimental Anesthesiology Sec- ty. The results indicate that cholinergic stimulation tion, Department of Anesthesiology and Intensive Care, Eberhard-Karls-Univer- reverses the depressant effects of sevoflurane by differ- sity, Schaffhausenstr. 113, D-72072 Tuebingen, Germany. [email protected]. Individual article reprints may be pur- ent mechanisms in cortical compared with spinal net- chased through the Journal Web site, www.anesthesiology.org. works (1) by increasing neuronal excitability in both

Anesthesiology, V 106, No 6, Jun 2007 1147 1148 GRASSHOFF ET AL. brain regions and (2) by decreasing anesthetic potency required to suppress movement in response to noxious stim- in cortical networks. ulation in 50% of subjects. We used the EC50 values for general anesthesia proposed by Franks and Lieb.30 Therefore, we assumed that 1 MAC corresponds to an aqueous concentra- 26 Materials and Methods tion of 0.35 mM sevoflurane. Sevoflurane was administered via bath perfusion using Spinal Slice Cultures gastight syringe pumps (ZAK Medicine Technique, Markt- All procedures were approved by the animal care com- heidenfeld, Germany), which were connected to the ex- mittee (Eberhard-Karls-University, Tuebingen, Germany) perimental chamber via Teflon tubing (Lee, Frankfurt, Ger- and were in accord with the German law on animal many). The flow rate was approximately 1 ml/min. To experimentation. Spinal cord slices were prepared from ensure steady state conditions, recordings during anes- embryos of pregnant 129/SvJ mice (days 14 and 15) Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 thetic treatment were conducted 10–15 min after starting according to the method described by Braschler et al.21 the perfusate change. The calibration of the recording Neocortical slices were obtained from 2- to 5-day-old system was performed as previously reported.31 129/SvJ mice as previously reported.22,23 Excised slices were placed on a coverslip and embedded Data Analysis in a plasma clot (Sigma, Taufkirchen, Germany). The cov- Data were bandpass filtered between 3 and 10 kHz as erslips were transferred into plastic tubes containing 0.75 acquired on a personal computer using the Digidata ml of nutrient fluid and incubated with 95% oxygen and 5% 1200 analog-to-digital/digital-to-analog interface (Axon carbon dioxide at 36.0°C for 1–2 h. One hundred milliliters Instruments, Union City, CA). Records were in addition of nutrient fluid consisted of 25 ml horse serum (Invitro- stored on a Sony data recorder PC 204A (Racal Elek- gen, Karlsruhe, Germany), 25 ml Hanks’ Balanced Salt So- tronik, Bergisch Gladbach, Germany). Further analysis lution (Sigma), and 50 ml Basal Medium Eagle (Sigma). For was performed using self-written software in OriginPro spinal slices, nutrient fluid included 10 nM Neuronal version 7 (OriginLab Corporation, Northampton, MA) Growth Factor (Sigma). The roller tube technique was used and MATLAB 6.5 (The MathWorks Inc., Natick, MA). to culture the tissue.24 After 1 day in culture, antimitotics Data analysis was conducted as described previously.26 (10 ␮M 5-fluoro-2-deoxyuridine, 10 ␮M cytosine-b-d-arabino- After close inspection of the data, a threshold was set furanoside, 10 ␮M uridine [all from Sigma]) were added to manually to avoid artifacts produced by baseline noise. The reduce proliferation of glial cells. Slices were used after 21 mean firing rate was obtained from single-unit or multiunit days in vitro for extracellular recordings. recordings and defined as the number of action potentials above threshold divided by the recording time of 180 s. In Extracellular Recordings addition to the effects of sevoflurane on mean firing rates, Spontaneous action potential activity was recorded as we normalized the data to eliminate acetylcholine-induced 13,25,26 reported previously. In spinal slices, extracellular changes of the basal activity. The data were normalized by recordings were performed from visually identified in- subtracting the firing rate monitored in the presence of 27–29 terneurons located in the ventral horn area. In brief, sevoflurane from the control rate (absence of the anes- slices were perfused with artificial cerebrospinal fluid thetic) for each experiment. The resulting difference was consisting of 120 mM NaCl, 3.3 mM KCl, 1.13 mM multiplied by 100 and divided by the control rate. Hence, a NaH2PO4,26mM NaHCO3, 1.8 mM CaCl2, and 11 mM value of 0% indicates sevoflurane lacked any inhibitory D-glucose. The artificial cerebrospinal fluid was bubbled drug action and a 100% depression indicates that not a with 95% oxygen and 5% carbon dioxide. Glass elec- single action potential occurred in the presence of the trodes with a resistance of approximately 2–5 M⍀ were anesthetic. filled with artificial cerebrospinal fluid and were intro- Results are given as mean Ϯ SEM. Concentration– duced into the tissue until extracellular spikes exceeding response curves were fitted by Hill equations using Orig- 100 ␮V in amplitude were visible and single-unit or inPro version 7. Estimated EC50 values were derived multiunit activity could be clearly identified. The noise from these fits. Statistical significance of differences be- (peak-to-peak) amplitude was usually approximately 50 tween concentration–response curves were assessed via ␮V. All experiments were performed at 34°–36°C. an F test, as previously reported.14 For statistical analysis between two data groups, the Student t test was used. Preparation and Application of Test Solutions Test solutions containing sevoflurane were obtained by dis- Results solving the liquid form of the anesthetic in artificial cerebrospinal fluid, which was equilibrated with 95% oxygen and 5% Opposing Actions of Sevoflurane and Acetylcholine carbon dioxide. A closed, air-free system was used to prevent on Action Potential Activity evaporation. Anesthetic levels are given as multiples of mini- The effects of sevoflurane and acetylcholine on action mum alveolar concentration (MAC) of an inhaled anesthetic potential activity monitored in neocortical slices are illus-

Anesthesiology, V 106, No 6, Jun 2007 CHOLINERGIC MODULATION OF SEVOFLURANE POTENCY 1149

Fig. 1. Effects of sevoflurane and acetyl- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 choline on action potential activity in neocortical slices. Sevoflurane data are depicted on the left side of the figure, and corresponding data obtained from experiments with acetylcholine are shown on the right side.(A) Original recordings are displayed in the absence (upper panel) and presence (lower panel) of 0.26 mM sevoflurane (corresponding to 0.75 MAC [1 MAC is defined to be the minimum alveolar concentration of an inhaled anesthetic required to suppress movement in response to noxious stimulation in 50% of subjects]) and 0.3 ␮M acetylcholine. Corresponding binned data are shown in B and C. Spikes were binned at 50-ms intervals. The depression of neuronal activity caused by 0.75 MAC sevoflurane arose as a decrease in the frequency of episodes of ongoing activity and a decrease in firing rates within these episodes. Although the episodes of ongoing activity were prolonged by the anesthetic, the overall time of neuronal silence was substantially lengthened (C, left side). Acetylcholine prolonged the duration of active states as observed with sevoflurane, but in contrast to the anesthetic, it increased average firing rates by elevating action potential activity within episodes of ongoing activity (C, right side). (D) Cumulative number of spikes over a recording period of 180 s.

trated in figure 1. Both substances altered neuronal firing lengthened. This pattern of sevoflurane effects on cortical patterns as anticipated: action potential activity was de- network activity corresponds largely to the pattern ob- creased by sevoflurane but increased by acetylcholine (fig. served with other ether anesthetics such as isoflurane and 1A). The depression of neuronal activity caused by sevoflu- enflurane, as reported previously.13 Acetylcholine pro- rane was displayed as a decrease in the frequency of bursts longed the duration of active states as observed with and a decrease in firing rates within these bursts (figs. 1B sevoflurane, but in contrast to the anesthetic, it increased and C). Although the bursts were prolonged by the anes- average firing rates by elevating action potential activity thetic, the overall time of neuronal silence was substantially within bursts (figs. 1C and D).

Anesthesiology, V 106, No 6, Jun 2007 1150 GRASSHOFF ET AL.

Fig. 2. Concentration-dependent effects of sevoflurane and acetylcholine in neocortical and spinal tissue slices. Levels of neuronal activity have been normalized to facilitate comparison. Sevoflurane reduced spontaneous activity of cortical (A) and spinal neurons (B) concentration-dependently within a range of clinically relevant concentrations. The anesthetic completely depressed neuronal activity at high concentrations in both preparations. Half-maximal effects were observed at 0.29 ؎ 0.01 mM in cortical and mM in spinal cultures corre- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 0.02 ؎ 0.17 sponding to 0.83 ؎ 0.03 MAC (1 MAC is defined to be the minimum alveolar concentration of an inhaled anesthetic required to suppress movement in response to noxious stimulation in 50% of -subjects) and 0.49 ؎ 0.05 MAC, respec tively. In contrast to sevoflurane, acetylcholine increased firing rates in both cortical (C) and spinal slices (D)upto twofold to threefold. Cholinergic enhancement of action potential activity was more pronounced in cortical compared with spinal slices.

Concentration-dependent effects of sevoflurane and ace- significantly increased action potential activity, demon- tylcholine in neocortical and spinal tissue slices are de- strating the existence of a cholinergic tone. The latter picted in figure 2. Levels of neuronal activity have been implication was corroborated by the finding that atro- normalized to facilitate comparison. Sevoflurane reduced pine reduced action potential firing of cortical and spinal spontaneous activity of cortical and spinal neurons concen- neurons by approximately 50%. These experiments im- tration-dependently within a range of clinically relevant ply that the increase in ongoing neuronal activity evoked concentrations. The anesthetic completely depressed neu- by bath applied acetylcholine (fig. 2) displays an effect ronal activity at high concentrations in both preparations. generated on top of a basal tone caused by intrinsically Half-maximal effects were observed at 0.29 Ϯ 0.01 mM in released acetylcholine. cortical and 0.17 Ϯ 0.02 mM in spinal cultures corresponding to 0.83 Ϯ 0.03 and 0.49 Ϯ 0.05 MAC, respectively. The Effects of Sevoflurane on Action Potential Activity EC value for depression of ongoing activity in spinal 50 in the Presence and Absence of Acetylcholine cultures obtained from mice turned out to be approxi- Concentration-dependent effects of sevoflurane in the mately 50% higher compared with the EC value reported 50 presence and absence of acetylcholine are presented in previously for rats (0.32 Ϯ 0.01 MAC).26 In contrast to figure 4. Acetylcholine augmented neuronal firing rates in sevoflurane, acetylcholine increased firing rates in both cortical slices at all sevoflurane concentrations tested, caus- cortical and spinal slices up to twofold to threefold. Cho- linergic enhancement of action potential activity was more pronounced in cortical compared with spinal slices. Satu- rating effects were observed between 10 and 100 ␮M acetylcholine in both preparations.

Intrinsic Release of Acetylcholine in Slice Cultures The results presented so far do not provide an answer to the question whether acetylcholine is synthesized and released in cortical and spinal slice cultures. Therefore, we investigated the effects of the reversible acetylcho- Fig. 3. Acetylcholine is released intrinsically in cortical and spinal tissue slices. The reversible acetylcholinesterase inhibi- linesterase inhibitor neostigmine as well as atropine, a tor neostigmine approximately doubled the firing rate at a competitive antagonist at muscarinic receptors, on on- concentration of 1 ␮M in cortical and spinal cultures, suggesting going neuronal activity. In these experiments, acetylcho- the existence of an intrinsic cholinergic release. This finding was corroborated by the observation that atropine, a competi- line was not added to the artificial cerebrospinal fluid. tive antagonist at muscarinic receptors, reduced action poten- The results are summarized in figure 3. Neostigmine tial firing in both preparations by approximately 50%.

Anesthesiology, V 106, No 6, Jun 2007 CHOLINERGIC MODULATION OF SEVOFLURANE POTENCY 1151

Fig. 4. Cholinergic modulation of sevoflurane effects on network activity in cortical and spinal slices. (A) Acetylcholine augmented neuronal firing rates in cortical slices at all sevoflurane concentrations tested, causing a concentration-dependent rightward and upward shift of the sevoflurane concentration–response curve. (B) In spinal slices, 10 ␮M acetylcholine produced a less pronounced increase in action potential activity. (C) Cholinergic modulation of normalized data of action potential activity depressed by sevoflurane in cortical slices. Acetyl-

Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 ؎ choline increased the EC50 from 0.83 0.03 MAC (1 MAC is defined to be the minimum alveolar concentration of an inhaled anesthetic required to suppress movement in response to noxious stimulation in 50% of subjects) in the absence of acetylcholine to 1.59 ؎ 0.32 MAC (1 ␮M acetylcholine) and 2.68 ؎ 0.57 MAC (10 ␮M acetylcholine). The concentration–response curves of sevoflurane measured in the presence of acetylcholine differed significantly from the concentration–response curve in the absence of acetylcholine (P < 0.001, F test). (D) Acetylcholine did not reduce the potency of sevoflurane in depressing spinal neurons (0.49 MAC in the absence compared with 0.05 ؎ .(MAC in the presence of acetylcholine). The concentration–response curves did not differ significantly (P > 0.1, F test 0.10 ؎ 0.47 ing a concentration-dependent rightward and upward shift and 2.68 Ϯ 0.57 MAC (10 ␮M acetylcholine) (fig. 4C and of the sevoflurane concentration response curve (fig. 4A). table 1). Both concentration–response curves of sevoflu- In spinal slices, enhancement of action potential activity rane measured in the presence of acetylcholine differed caused by 10 ␮M acetylcholine appeared less pronounced significantly from the concentration–response curve in the compared with cortical slices (fig. 4B). absence of acetylcholine (P Ͻ 0.001, F test). In contrast to Further interactions between acetylcholine and sevoflu- neocortical networks, acetylcholine did not reduce the rane were characterized by normalizing the changes in potency of sevoflurane in depressing spinal neurons (0.49 neuronal activity caused by the anesthetic in the absence Ϯ 0.05 MAC in the absence compared with 0.47 Ϯ 0.10 and presence of acetylcholine (see Materials and Methods). MAC in the presence of acetylcholine; fig. 4D and table 1). The plots in figures 4C and D show the concentration- The concentration–response curves did not differ signifi- dependent effects of sevoflurane ranging to 100% inhibi- cantly (P Ͼ 0.1, F test). tion, corresponding to a total depression of firing rates in both preparations. In cortical slices, acetylcholine attenu- Effects of Sevoflurane in Cortical and Spinal Slices ated sevoflurane potency in a concentration-dependent in the Presence of Bicuculline Ϯ manner. The EC50 was increased from 0.83 0.03 MAC (0 How can we explain the finding that acetylcholine ␮M acetylcholine) to 1.59 Ϯ 0.32 MAC (1 ␮M acetylcholine) reduces sevoflurane potency in cortical but not in spinal

Table 1. Parameters of Sevoﬂurane Concentration–Response Curves Measured in the Absence and Presence of Acetylcholine in Cultured Cortical and Spinal Slices

2 EC50 Hill Coefﬁcient Goodness of Fit (R )

Cortical Sevoflurane 0.83 Ϯ 0.03 MAC (0.29 Ϯ 0.01 mM) 1.49 Ϯ 0.06 0.999 Sevoflurane ϩ acetylcholine, 1 ␮M 1.59 Ϯ 0.32 MAC (0.56 Ϯ 0.11 mM) 1.33 Ϯ 0.18 0.989 Sevoflurane ϩ acetylcholine, 10 ␮M 2.68 Ϯ 0.57 MAC (0.94 Ϯ 0.20 mM) 1.19 Ϯ 0.12 0.998 Sevoflurane ϩ bicuculline, 1 ␮M 1.44 Ϯ 0.26 MAC (0.50 Ϯ 0.09 mM) 1.51 Ϯ 0.40 0.986 Spinal Sevoflurane 0.49 Ϯ 0.05 MAC (0.17 Ϯ 0.02 mM) 1.30 Ϯ 0.17 0.984 Sevoflurane ϩ acetylcholine, 10 ␮M 0.37 Ϯ 0.10 MAC (0.13 Ϯ 0.04 mM) 0.92 Ϯ 0.30 0.934 Sevoflurane ϩ bicuculline, 1 ␮M 0.42 Ϯ 0.05 MAC (0.15 Ϯ 0.02 mM) 1.18 Ϯ 0.35 0.979

Half-maximal depression of average spike rates and Hill coefﬁcients were calculated from the concentration–response curves displayed in ﬁgures 4 and 5 . The

EC50 values are provided in millimolars and minimum alveolar concentration (MAC). 1 MAC is deﬁned to be the minimum alveolar concentration of an inhaled anaesthetic required to suppress movement in response to noxious stimulation in 50% of subjects.

Anesthesiology, V 106, No 6, Jun 2007 1152 GRASSHOFF ET AL.

neurons? We hypothesized that this difference is related to distinct molecular targets of sevoﬂurane in cortical and spinal networks as well as to acetylcholine-induced

changes in the efﬁcacy of GABAA receptor–mediated inhibition. In particular, we hypothesized that sevoﬂu- rane depresses neuronal activity in cortical networks

predominantly via enhancing GABAA receptor function, as demonstrated previously for isoflurane and enflu- 13 rane. In contrast to the neocortex, GABAA receptors are only a minor target of sevoflurane in the spinal cord.26,32,33 Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 This hypothesis was tested as follows: Quantitative effects of acetylcholine were mimicked in spinal and neocortical slices by applying bicuculline, a specific

GABAA receptor antagonist. Thereby, two effects exerted by acetylcholine were simulated, namely an increase in neuronal activity and a decrease in the efﬁcacy

of GABAA receptor–mediated inhibition. According to the hypothesis proposed in the last paragraph, bicuculline is expected to decrease sevoflurane potency in cortical but not in spinal slices. A concentration of 1 ␮M bicuculline was used in the experiments because it approximately doubled ongoing neuronal activity in cultured cortical and spinal slices (fig. 5A) and can conse- quently be regarded as equipotent with regard to the effects on network activity induced by acetylcholine. In neocortical slices, bicuculline induced an increase in Ϯ Ϯ EC50 of sevoflurane from 0.83 0.03 to 1.44 0.26 MAC (fig. 5B). Both concentration–response curves dif- Ͻ fered significantly (P 0.01, F test). In contrast to the Fig. 5. Bicuculline effects on the potency of sevoflurane in neocortex, bicuculline did not significantly reduce the depressing neuronal network activity in cortical and spinal potency of sevoflurane in depressing spinal neurons slices. (A) Action potential activity was approximately doubled Ϯ Ϯ by 1 ␮M bicuculline in both preparations in the absence of ,B) In neocortical slices) .(10 ؍ MAC in the absence compared with 0.42 sevoflurane (* P < 0.05, t test, n 0.05 0.49) ␮ 0.05 MAC in the presence of bicuculline; fig. 5C). In 1 M bicuculline induced an increase in EC50 of sevoflurane from 0.83 ؎ 0.03 MAC (1 MAC is defined to be the minimum accord with the EC50 values, the concentration–re- Ͼ alveolar concentration of an inhaled anesthetic required to sponse curves were not different (P 0.1, F test). suppress movement in response to noxious stimulation in 50% In the following step, we investigated whether the of subjects) to 1.44 ؎ 0.26 MAC. Both concentration–response depressant effects of sevoflurane in neocortical slices curves differed significantly (P < 0.01, F test). (C) In contrast to neocortical slices, bicuculline did not affect the potency of can be completely antagonized by higher bicuculline sevoflurane in depressing spinal neurons (0.49 ؎ 0.05 MAC in concentrations (100 ␮M) as reported previously for other the absence compared with 0.42 ؎ 0.05 MAC in the presence of 34 ether derivates. A concentration of 100 ␮M bicuculline bicuculline). In accord with the EC50 values, the concentration– response curves were not different (P > 0.1, F test). abolished the depression of neuronal network activity by 0.26 mM (corresponding to 0.75 MAC) sevoflurane almost completely (from 49.61 Ϯ 3.41% in the absence to spinal networks. The results displayed in figure 5 sup- 2.87 Ϯ 5.13% in the presence of bicuculline; P Ͻ 0.001, port the proposed mechanism that acetylcholine atten- n ϭ 10). This result points to enhanced GABAergic uates the impact of GABAergic inhibition. To corrobo- synaptic inhibition being the predominant mechanism rate this hypothesis, the effects of etomidate, an almost mediating the depressant effects of 0.75 MAC sevoflu- selective modulator at GABAA receptors, were investi- rane in neocortical slices (fig. 6). gated. The selectivity of etomidate at GABAA receptors has been demonstrated previously for both cortical and 35,36 Effects of Acetylcholine on Etomidate Potency in spinal networks in vitro and in vivo. Therefore, the Cortical and Spinal Slices hypothesis to be tested was that acetylcholine reduces In the last set of experiments, the interactions be- the relative inhibition of neuronal activity exerted by tween sevoflurane and bicuculline were studied to elu- etomidate not only in cortical but also in spinal slices.

cidate how a decrease in GABAA receptor–mediated in- Effects of etomidate on cortical and spinal neurons were hibition alters the potency of sevoﬂurane in cortical and investigated at a concentration of 1.5 ␮M. This concen-

Anesthesiology, V 106, No 6, Jun 2007 CHOLINERGIC MODULATION OF SEVOFLURANE POTENCY 1153

Moreover, interactions of reversible and irreversible blockers of acetylcholinesterase with muscarinic receptors affect GABAergic transmission in hippocampal neurons in vitro.41 But how will a decrease in GABA release alter anesthetic mechanisms? We predict that the potency of GABAergic anesthetics in depressing neuronal network activity is quantitatively linked to the extracellular GABA concentration. This hypothesis is based on the fact that at clinically relevant concentrations, anesthetics frequently Fig. 6. Effects of 0.75 MAC sevoﬂurane (1 MAC is deﬁned to be the minimum alveolar concentration of an inhaled anesthetic act as positive modulators at GABAA receptors. Therefore,

required to suppress movement in response to noxious stimu- they require the presence of GABA at the agonist binding Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 lation in 50% of subjects) in the absence and presence of 100 ␮M 42–44 site for affecting GABAA receptor function. The exper- bicuculline. A concentration of 100 ␮M bicuculline abolished the depression of neuronal network activity by 0.26 mM (corre- imental finding that acetylcholine impairs the potency of sponding to 0.75 MAC) sevoflurane almost completely (from sevoflurane in decreasing neuronal activity in cortical slices in the absence to 2.87 ؎ 5.13% in the presence of 3.41% ؎ 49.61 This result points to is therefore well explained by a modulatory action of .(10 ؍ bicuculline; *** P < 0.001, t test, n enhanced synaptic inhibition mediated via ␥-aminobutyric acid sevoflurane at GABAA receptors. type A receptors being the predominate mechanism to cause Unlike in neocortex, sevoflurane potency remained the depressant effects of 0.75 MAC sevoflurane in neocortical slices. unaffected by acetylcholine in spinal cord slices. What is the reason for this brain region–specific difference? tration decreased neuronal activity by approximately In the current study, we have provided evidence that 60% in both cortical and spinal networks in the absence the GABAA receptor is a major target for sevoflurane in of acetylcholine. In cortical slices, the relative inhibition neocortical microcircuits. However, we recently dem- in the presence of acetylcholine was reduced from 63.4 onstrated that in cultured spinal slices sevoflurane Ϯ 6.8 to 29.7 Ϯ 13.7% (n ϭ 10, P Ͻ 0.05), whereas depresses neuronal activity via multiple molecular ␮ acetylcholine decreased the relative inhibition of 1.5 M targets and that modulation of GABAA receptors is not etomidate to depress spinal neurons from 65.5 Ϯ 3.9% to the major mechanism by which sevoflurane decreases 47.0 Ϯ 4.1% (n ϭ 8, P Ͻ 0.01). The results clearly the excitability of spinal neurons.26 The different ac- demonstrate that in contrast to sevoflurane, the relative tions of sevoflurane on neocortical and spinal net- inhibition of etomidate was reduced not only in cortical works are clearly shown in figure 5. The specific but also in spinal networks, supporting the hypothesis GABAA receptor antagonist bicuculline increases neu- that acetylcholine attenuates the impact of GABAergic ronal basal activity at a concentration of 1 ␮M in inhibition in both preparations. neocortical and spinal slices by approximately twofold, indicating that in both networks action potential Discussion firing is under substantial GABAergic control. How- ever, bicuculline reduces sevoflurane potency only in Mechanisms Underlying Cholinergic-induced neocortical slices, supporting the hypothesis that ace- Decrease in Sevoflurane Potency tylcholine affects sevoflurane potency just minimally The cholinergic system is an important modulatory in spinal slices because GABA receptors are not a neurotransmitter system in the brain.37 Activation of A predominant target of the anesthetic. cholinergic innervation of the cortex has been impli- However, the finding that acetylcholine did not affect the cated in sensory processing, learning, and memory. The potency of sevoflurane in depressing action potential firing application of the reversible acetylcholinesterase inhibi- of spinal slices can alternatively be explained by assuming tor tacrine has been demonstrated to promote plasticity and learning in the motor cortex of healthy volunteers.38 that acetylcholine does not affect GABAA receptor–medi- At the cellular level, high concentrations of the cholin- ated inhibition in the spinal cord. To rule out this possibil- ergic agonist carbachol have been shown to have a ity, interactions between acetylcholine and the almost se- strong desynchronizing action on neuronal activity mea- lective GABAA receptors modulator etomidate were sured in cultured cortical neurons.39 In layer V of the rat investigated, hypothesizing that acetylcholine did not affect visual cortex, acetylcholine both increases excitability etomidate potency in spinal slices. The hypothesis had to and depresses synaptic transmission by affecting differ- be abandoned because acetylcholine decreased the po- ent GABAergic interneurons via distinct cholinergic re- tency of etomidate in both spinal and cortical networks. ceptors.40 Furthermore, there is ample evidence that The finding that acetylcholine significantly affects etomi- cholinergic stimulation decreases GABA release in corti- date potency in spinal slices clearly argues against the

cal networks by activating muscarinic and nicotinic re- possibility that acetylcholine does not modulate GABAA ceptors located on presynaptic terminals.9,10 receptor–mediated inhibition in the spinal cord.

Anesthesiology, V 106, No 6, Jun 2007 1154 GRASSHOFF ET AL.

Brain Acetylcholine Concentration during as diazepam, which are potent modulators of GABAA Cholinergic Crisis receptors, are recommended for acute treatment of ep- The brain acetylcholine concentrations occurring dur- ileptiform seizure activity in intoxicated patients.7,49 In ing a severe organophosphate intoxication are hard to rats, diazepam interrupts soman-induced seizures only measure in vivo. However, in an outstanding study, when applied within the first 5–10 min after seizure Tonduli et al.3 exposed freely moving rats to soman and onset.50 It seems likely that a severe decrease in the acquired three sets of neurophysiologic data before and potency of diazepam as caused by massive cholinergic during the intoxication. They determined cortical acetyl- overstimulation may attribute to this temporally limited cholinesterase activity and acetylcholine concentrations effect. by microdialysis and associated the parameters with The conclusion that cholinergic stimulation increases electroencephalographic recordings as well as with drug requirement for providing anesthesia is also corrob- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 power spectrum analysis of the ␥ band. Acetylcholine orated by human studies using a reversible inhibitor of levels measured in 60-␮l microdialysis fractions obtained acetylcholinesterase. Meuret et al.47 investigated the ef- from intoxicated rats ranged from 40 Ϯ 8 pmol in sub- fects of physostigmine in human subjects anesthetized jects showing no seizure activity to 102 Ϯ 21 pmol in with propofol by assessing central nervous system func- rats displaying severe seizure activity,3 corresponding to tion by use of the Auditory Steady State Response and concentrations of 0.7 and 1.7 ␮M in the perfusate of the Bispectral Index. Physostigmine restored consciousness fraction, respectively. The concentration in extracellular with concomitant increases in the Auditory Steady State space is always higher compared with the perfusate Response and Bispectral Index. Using a similar experi- concentration, depending on the type of probe and the mental approach, Plourde et al.51 report that sevoflurane perfusion speed. In the study performed by Tonduli et anesthesia is also antagonized by physostigmine. al., the microdialysis probes were perfused with 2 ␮l/ In summary, the studies on the effects of anticholines- min, and a fraction was collected every 30 min. With this terases on experimental animals and human subjects perfusion speed, a relative recovery of 20% can be as- clearly indicate that cholinergic overstimulation may sumed, leading to concentrations between 3.5 and 8.5 considerably increase drug requirement for providing ␮M in the extracellular space depending on the degree of general anesthesia. However, raising anesthetic concen- soman intoxication. Therefore, the concentrations of 1 trations into a high-dose range in patients with organo- and 10 ␮M used in our experiments cover well the range phosphorus intoxication may not always be an option, of extracellular acetylcholine concentrations observed in because these patients frequently display cardiac abnor- intoxicated rats. malities and hemodynamic instability.1,7 An alternative treatment may be the coapplication of anesthetics and Cholinergic Modulation of Drug Requirement for antagonists of muscarinic acetylcholine receptors such Providing Anesthesia as atropine. The rational for this option is that acetylcho- In an elegant study, Hudetz et al.45 recently showed line-induced reversal of hypnosis is largely mediated via that neostigmine, a reversal blocker of acetylcholinest- muscarinic receptors. Hudetz et al.45 demonstrated that erase, inverts isoflurane-induced hypnosis in rats. In ad- muscarinic agonists mimic the effects of anticholinester- dition to these results, there is multiple evidence that ases on the righting reflex and the electroencephalo- cholinesterase blockers reverse anesthesia by increasing gram. This finding is corroborated by the observation of brain acetylcholine levels.46,47 In the current study, we Douglas et al.52,53 that cholinergic electroencephalo- identified two putative mechanisms: First, we observed graphic arousal is largely mediated by M1 and M2 recep- that acetylcholine increases neuronal basal activity in tors. Without blocking muscarinic receptors, anesthesi- spinal and cortical slices by twofold to threefold (fig. 2). ologists may be unable to determine anesthetic Because general anesthesia is based on a severe depres- requirement because acetylcholine concentrations in sion of central nervous functions, the excitatory actions the central nervous system are unknown and may un- exerted by acetylcholine must be counterbalanced by dergo gross changes in patients with organophosphorus increasing anesthetic concentrations. Second, we intoxication. However, more direct studies to test these showed that acetylcholine compromises anesthetic po- ideas in whole animals are required before recommen- tency, a finding that is probably explained by an impair- dations for approaches to clinical care can be made. ment of GABAA receptor–mediated inhibition. Both mechanisms, namely the elevation of neuronal basal ac- The authors thank Claudia Holt (Technical Assistant, Eberhard-Karls-Univer- sity, Tuebingen, Germany) for excellent technical assistance. tivity (fig. 4) and the decrease in anesthetic potency, may significantly increase drug requirement for providing general anesthesia in patients with organophosphate in- References toxication. The pharmacologic management of organo- 1. Ben Abraham R, Rudick V, Weinbroum AA: Practical guidelines for acute phosphate poisoning has extensively been discussed by care of victims of bioterrorism: Conventional injuries and concomitant nerve 48,49 Lallement et al. Remarkably, benzodiazepines such agent intoxication. ANESTHESIOLOGY 2002; 97:989–1004

Anesthesiology, V 106, No 6, Jun 2007 CHOLINERGIC MODULATION OF SEVOFLURANE POTENCY 1155

2. Lallement G, Denoyer M, Collet A, Pernot MI, Baubichon D, Monmaur P, 29. Ballerini L, Galante M: Network bursting by organotypic spinal slice Blanchet G: Changes in hippocampal acetylcholine and glutamate extracellular cultures in the presence of bicuculline and/or strychnine is developmentally levels during soman-induced seizures: Influence of septal cholinoceptive cells. regulated. Eur J Neurosci 1998; 10:2871–9 Neurosci Lett 1992; 139:104–7 30. Franks NP, Lieb WR: Molecular and cellular mechanisms of general anaes- 3. Tonduli LS, Testylier G, Marino IP, Lallement G: Triggering of soman- thesia. Nature 1994; 367:607–14 induced seizures in rats: Multiparametric analysis with special correlation be- 31. Antkowiak B, Heck D: Effects of the volatile anesthetic enflurane on tween enzymatic, neurochemical and electrophysiological data. J Neurosci Res spontaneous discharge rate and GABA(A)-mediated inhibition of Purkinje cells in 1999; 58:464–73 rat cerebellar slices. J Neurophysiol 1997; 77:2525–38 4. de Jong RH: Nerve gas terrorism: A grim challenge to anesthesiologists. 32. Zhang Y, Wu S, Eger EI, Sonner JM: Neither GABA(A) nor strychnine- Anesth Analg 2003; 96:819–25 sensitive glycine receptors are the sole mediators of MAC for isoflurane. Anesth 5. Cosar A, Kenar L: An anesthesiological approach to nerve agent victims. Mil Analg 2001; 92:123–7 Med 2006; 171:7–11 33. Zhang Y, Sonner JM, Eger EI, Stabernack CR, Laster MJ, Raines DE, Harris 6. Eddleston M, Phillips MR: Self poisoning with pesticides. BMJ 2004; 328: RA: Gamma-aminobutyric acid A receptors do not mediate the immobility pro- 42–4 duced by isoflurane. Anesth Analg 2004; 99:85–90 7. White SM: Chemical and biological weapons: Implications for anaesthesia 34. Antkowiak B, Hentschke H, Kirschfeld K: Effects of volatile anaesthetics and intensive care. Br J Anaesth 2002; 89:306–24 on spontaneous action potential firing of cerebellar Purkinje cells in vitro do not 8. Baker DJ: Management of casualties from terrorist chemical and biological Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/106/6/1147/364219/0000542-200706000-00015.pdf by guest on 01 October 2021 follow the Meyer-Overton rule. Br J Anaesth 1997; 79:617–24 attack: A key role for the anaesthetist. Br J Anaesth 2002; 89:211–4 35. Grasshoff C, Jurd R, Rudolph U, Antkowiak B: Spinal ␤3-containing GABA 9. Hashimoto T, Shu H, Kuriyama K: Muscarinic M1 receptor mediated inhi- A receptors mediate pre- and postsynaptic effects of etomidate (abstract). ANESTHE- bition of GABA release from rat cerebral cortex. Neurochem Int 1994; 24:389–94 SIOLOGY 2006; 105:A1115 10. Yamazaki Y, Jia Y, Hamaue N, Sumikawa K: Nicotine-induced switch in the nicotinic cholinergic mechanisms of facilitation of long-term potentiation induc- 36. Jurd R, Arras M, Lambert S, Drexler B, Siegwart R, Crestani F, Zaugg M, tion. Eur J Neurosci 2005; 22:845–60 Vogt KE, Ledermann B, Antkowiak B, Rudolph U: General anesthetic actions in 11. Kohling R, Vreugdenhil M, Bracci E, Jefferys JG: Ictal epileptiform activity vivo strongly attenuated by a point mutation in the GABA(A) receptor beta3 is facilitated by hippocampal GABAA receptor-mediated oscillations. J Neurosci subunit. FASEB J 2003; 17:250–2 2000; 20:6820–9 37. Perry E, Walker M, Grace J, Perry R: Acetylcholine in mind: A neurotrans- 12. Thompson SM, Gahwiler BH: Activity-dependent disinhibition: I. Repeti- mitter correlate of consciousness? Trends Neurosci 1999; 22:273–80 tive stimulation reduces IPSP driving force and conductance in the hippocampus 38. Korchounov A, Ilic TV, Schwinge T, Ziemann U: Modification of motor in vitro. J Neurophysiol 1989; 61:501–11 cortical excitability by an acetylcholinesterase inhibitor. Exp Brain Res 2005; 13. Antkowiak B, Helfrich FC: Effects of small concentrations of volatile 164:399–405 anesthetics on action potential firing of neocortical neurons in vitro.ANESTHESI- 39. Tateno T, Jimbo Y, Robinson HP: Spatio-temporal cholinergic modulation OLOGY 1998; 88:1592–605 in cultured networks of rat cortical neurons: Spontaneous activity. Neuroscience 14. Hentschke H, Schwarz C, Antkowiak B: Neocortex is the major target of 2005; 134:425–37 sedative concentrations of volatile anaesthetics: Strong depression of firing rates 40. Xiang Z, Huguenard JR, Prince DA: Cholinergic switching within neocor- and increase of GABA receptor-mediated inhibition. Eur J Neurosci 2005; 21:93– tical inhibitory networks. Science 1998; 281:985–8 102 41. Santos MD, Pereira EF, Aracava Y, Castro NG, Fawcett WP, Randall WR, 15. Fiset P, Paus T, Daloze T, Plourde G, Meuret P, Bonhomme V, Hajj AN, Albuquerque EX: Low concentrations of pyridostigmine prevent soman-induced Backman SB, Evans AC: Brain mechanisms of propofol-induced loss of conscious- inhibition of GABAergic transmission in the central nervous system: Involvement ness in humans: A positron emission tomographic study. J Neurosci 1999; of muscarinic receptors. J Pharmacol Exp Ther 2003; 304:254–65

19:5506–13 42. Li X, Czajkowski C, Pearce RA: Rapid and direct modulation of GABAA 16. Grasshoff C, Rudolph U, Antkowiak B: Molecular and systemic mecha- receptors by halothane. ANESTHESIOLOGY 2000; 92:1366–75 nisms of general anaesthesia: The “multi-site and multiple mechanisms” concept. 43. Li X, Pearce RA: Effects of halothane on GABA(A) receptor kinetics: Curr Opin Anaesthesiol 2005; 18:386–91 Evidence for slowed agonist unbinding. J Neurosci 2000; 20:899–907 17. Rudolph U, Antkowiak B: Molecular and neuronal substrates for general 44. Lin LH, Chen LL, Zirrolli JA, Harris RA: General anesthetics potentiate anaesthetics. Nat Rev Neurosci 2004; 5:709–20 gamma-aminobutyric acid actions on gamma-aminobutyric acid A receptors ex- 18. Collins JG, Kendig JJ, Mason P: Anesthetic actions within the spinal cord: pressed by Xenopus oocytes: Lack of involvement of intracellular calcium. Contributions to the state of general anesthesia. Trends Neurosci 1995; 18: J Pharmacol Exp Ther 1992; 263:569–78 549–53 45. Hudetz AG, Wood JD, Kampine JP: Cholinergic reversal of isoflurane 19. Campagna JA, Miller KW, Forman SA: Mechanisms of actions of inhaled anesthesia in rats as measured by cross-approximate entropy of the electroen- anesthetics. N Engl J Med 2003; 348:2110–24 cephalogram. ANESTHESIOLOGY 2003; 99:1125–31 20. Sonner JM, Antognini JF, Dutton RC, Flood P, Gray AT, Harris RA, Homan- 46. Fassoulaki A, Sarantopoulos C, Derveniotis C: Physostigmine increases the ics GE, Kendig J, Orser B, Raines DE, Trudell J, Vissel B, Eger EI: Inhaled dose of propofol required to induce anaesthesia. Can J Anaesth 1997; 44:1148–51 anesthetics and immobility: Mechanisms, mysteries, and minimum alveolar anes- 47. Meuret P, Backman SB, Bonhomme V, Plourde G, Fiset P: Physostigmine thetic concentration. Anesth Analg 2003; 97:718–40 reverses propofol-induced unconsciousness and attenuation of the auditory 21. Braschler UF, Iannone A, Spenger C, Streit J, Luscher HR: A modified roller steady state response and Bispectral Index in human volunteers. ANESTHESIOLOGY tube technique for organotypic cocultures of embryonic rat spinal cord, sensory 2000; 93:708–17 ganglia and skeletal muscle. J Neurosci Methods 1989; 29:121–9 48. Lallement G, Dorandeu F, Filliat P, Carpentier P, Baille V, Blanchet G: 22. Drexler B, Roether CL, Jurd R, Rudolph U, Antkowiak B: Opposing actions Medical management of organophosphate-induced seizures. J Physiol Paris 1998; of etomidate on cortical theta oscillations are mediated by different ␥-aminobu- 92:369–73 tyric acid type A receptor subtypes. ANESTHESIOLOGY 2005; 102:346–52 23. Drexler B, Jurd R, Rudolph U, Antkowiak B: Dual actions of enflurane on 49. Lallement G, Baubichon D, Clarencon D, Galonnier M, Peoc’h M, Carpen- postsynaptic currents abolished by the ␥-aminobutyric acid type A receptor tier P: Review of the value of gacyclidine (GK-11) as adjuvant medication to conventional treatments of organophosphate poisoning: Primate experiments ␤3(N265M) point mutation. ANESTHESIOLOGY 2006; 105:297–304 24. Gahwiler BH: Organotypic monolayer cultures of nervous tissue. J Neuro- mimicking various scenarios of military or terrorist attack by soman. Neurotoxi- sci Methods 1981; 4:329–42 cology 1999; 20:675–84 25. Antkowiak B: Different actions of general anesthetics on the firing patterns 50. Shih T, McDonough-JH J, Koplovitz I: Anticonvulsants for soman-induced seizure activity. J Biomed Sci 1999; 6:86–96 of neocortical neurons mediated by the GABAA receptor. ANESTHESIOLOGY 1999; 91:500–11 51. Plourde G, Chartrand D, Fiset P, Font S, Backman SB: Antagonism of 26. Grasshoff C, Antkowiak B: Propofol and sevoflurane depress spinal neu- sevoflurane anaesthesia by physostigmine: Effects on the auditory steady-state rons in vitro via different molecular targets. ANESTHESIOLOGY 2004; 101:1167–76 response and bispectral index. Br J Anaesth 2003; 91:583–6 27. Streit J, Spenger C, Luscher HR: An organotypic spinal cord–dorsal root 52. Douglas CL, Baghdoyan HA, Lydic R: Postsynaptic muscarinic M1 recep- ganglion–skeletal muscle coculture of embryonic rat: II. Functional evidence for tors activate prefrontal cortical EEG of C57BL/6J mouse. J Neurophysiol 2002; the formation of spinal reflex arcs in vitro. Eur J Neurosci 1991; 3:1054–68 88:3003–9 28. Spenger-C, Braschler UF, Streit J, Luscher HR: An organotypic spinal 53. Douglas CL, Baghdoyan HA, Lydic R: Prefrontal cortex acetylcholine re- cord–dorsal root ganglion skeletal muscle coculture of embryonic rat: I. The lease, EEG slow waves, and spindles are modulated by M2 autoreceptors in morphological correlates of the spinal reflex arc. Eur J Neurosci 1991; 3:1037–53 C57BL/6J mouse. J Neurophysiol 2002; 87:2817–22

Anesthesiology, V 106, No 6, Jun 2007