GABAergic mechanisms in the action of general anesthetics

Toxicology Letters 100 – 101 (1998) 203 – 207

GABAergic mechanisms in the action of general anesthetics K. Hirota a,*, S.H. Roth b, J. Fujimura a, A. Masuda a, Y. Ito a a

Department of Anesthesiology, Toyama Medical and Pharmaceutical Uni6ersity School of Medicine, 2630 Sugitani, Toyama 930 -01, Japan b Departments of Pharmacology & Therapeutics and Anaesthesia, Faculty of Medicine, Uni6ersity of Calgary, 3330 Hospital Dri6e, N.W. Calgary, Alta. T2N 4N1, Canada Accepted 7 May 1998

Abstract 1. The effects of volatile and intravenous anesthetics were studied on evoked field potentials in rat hippocampal CA1 neurons in vitro to determine the role of GABAergic mechanisms in the action of general anesthetics. 2. It was observed that both volatile (halothane, isoflurane, sevoflurane) and intravenous (thiopental, pentobarbital, propofol) anesthetics decreased population spike (PS) amplitudes. 3. Using paired-pulse paradigms, it was revealed that volatile agents enhance paired-pulse facilitation (PPF), and intravenous agents reduce PPF. Use-dependent effects on PS amplitudes were observed following application of the intravenous anesthetics, whereas volatile agents did not show use-dependency. The effects of the intravenous anesthetics were blocked by the GABAA receptor antagonist, bicuculline. 4. It is suggested that agent specific actions of general anesthetics are a result of differential effects on GABAergic mechanisms that modulate synaptic transmission. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Agent specific action; General anesthetics; Population spike; Rat hippocampal slice

1. Introduction It has been proposed that general anesthetics reduce neural excitability as a result of a common mechanism such as GABA-mediated inhibition

* Corresponding author. Tel.: + 81 764 342281 (extn. 3047); fax: +81 764 345040; e-mail: [email protected]

(Higashi and Nishi, 1982; Nakahiro et al., 1989). In contrast to a ’unitary’ theory, many studies have demonstrated that general anesthetics can produce agent and site specific effects in the central nervous system (MacIver and Roth, 1988; MacIver and Kendig, 1991). In the present study, we have compared the effects of selected volatile and intravenous anesthetics on both pre- and post-synaptic GABAergic mechanisms using the in vitro rat hippocampal preparation. The results

0378-4274/98/$ - see front matter © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0378-4274(98)00186-6

204

K. Hirota et al. / Toxicology Letters 100–101 (1998) 203–207

suggest that agent specific actions of general anesthetics occur as a result of differential effects on GABAergic mechanisms which modulate synaptic transmission in the central nervous system (CNS).

2. Materials and methods The hippocampal slice preparation was prepared as previously described (Hirota and Roth, 1997a,b). Ethical approval was obtained from the Animal Care Committee of the Toyama Medical and Pharmaceutical University. Male Wister rats (100 –200 g) were anesthetized with sevoflurane (5.0 vol.%) in oxygen. The brain was rapidly removed and placed in cold artificial cerebrospinal fluid (ACSF). Composition of the ACSF was (mM): NaCl 124, KCl 5, CaCl2 2, NaH2PO4 1.25, MgSO4 2, NaHCO3 26, glucose 10. The ACSF was pre-cooled (8 – 10°C) and saturated with 95%O2/5%CO2 before use (pH 7.3 – 7.4). Transverse slices (400 mm) of the dissected hippocampus were prepared using a Rotorslicer DTY-7700 (DSK, Osaka, Japan), and placed on a nylon mesh screen at the gas/liquid interface in a recording chamber. The chamber was continuously perfused (1.5 ml/min) with oxygenated and prewarmed (37°C) ACSF. The upper surface of the slices were exposed to humidified 95%O2/ 5%CO2 gas mixture. Slices were incubated for 90 min without electrical stimulation. Bipolar nichrome stimulating electrodes were placed on Schaffer collateral fibers, and a glass microelectrode (3–5 MV; 2 M NaCl) was placed in the cell body regions of CA1 for recording extracellular field potentials. Stimuli (0.05 ms duration, 0.1 Hz) were delivered by a Nihon Koden (Tokyo, Japan) SEN-3301 stimulator. Field potentials were amplified (Nihon Koden MEZ-8201) with low- and high-frequency filters set at 1 Hz and 10 kHz respectively. The signal was digitally converted (14 kHz) using a MacADIOS system (GWI, Somerville, MA, USA) and stored on a Macintosh computer (Apple, Cupertino, CA, USA) for later analysis. Paired-pulse stimuli (train of two identical stimuli) were delivered to generate

paired-pulse facilitation (PPF) of field population spikes (PS). The minimum stimulus intensity (5– 10 V) that elicited a maximum PS amplitude was used. PS amplitudes were measured from peak positive to peak negative. Volatile anesthetics were applied to the tissue chamber as a vapour via the pre-warmed, humidified and oxygenated gas stream above the slices using a calibrated vaporizer. Concentrations, expressed as volume percent (vol.%), refer to dial settings on the vaporizer. Concentrations of volatile anesthetics in ACSF were determined using gas chromatography. Intravenous agents were directly dissolved in ACSF except for propofol which was first dissolved in Intralipid™ and then diluted into ACSF. All preparations used exhibited PS amplitudes of at least 5 mV and control variability less than 5% during the initial data acquisition period and following washout of anesthetic. Results were expressed as mean9 S.D. Differences among multiple groups were tested by ANOVA, and differences between paired sets of data were compared by two-tailed t-test. A P value of B 0.05 was considered significantly different.

Fig. 1. Representative example of the effects of volatile anesthetic isoflurane (2.0 vol.%) and intravenous anesthetic thiopental (2 ×10 − 4 M) on the evoked population spikes (PSs) of hippocampal CA1 neurons in the absence (upper trace) and presence (lower trace) of bicuculline methiodide (5× 10 − 5 M). Field PSs were elicited with a paired-pulse stimulus (3 – 8 V, 0.05 ms, 0.1 Hz) with a interstimulus interval of 35 ms.

K. Hirota et al. / Toxicology Letters 100–101 (1998) 203–207

205

Fig. 2. The effects of isoflurane (2.0 vol.%) and thiopental (2 × 10 − 4 M) on the pairedpulse facilitation of population spikes in the absence (upper trace) and presence (lower trace) of bicuculline methiodide (5× 10 − 5 M). The ratio for the second/first evoked population spike amplitudes (PS2:PS1) was plotted as a function of the interstimulus intervals (5 – 1000 ms). Open symbols, control; closed symbols, anesthetic. Each symbol represents a mean value from three to five slices. *, PB 0.05 by two-way ANOVA.

3. Results Fig. 1 shows example recordings of the volatile anesthetic, isoflurane (2.0 vol.%), and intravenous anesthetic, thiopental (2×10 − 4 M), on the evoked PSs of CA1 pyramidal neurons in the absence and presence of the GABAA antagonist, bicuculline methiodide (BMI; 5× 10 − 5 M). In the absence of BMI, both anesthetics reduced the PS amplitudes in a reversible manner. Isoflurane had a greater effect on the first evoked PS (PS1) than on the second PS (PS2), in contrast to thiopental which was more effective on PS2 than PS1. The effects of thiopental on PS were completely antag-

onized with BMI; however, BMI failed to completely antagonize the effects of isoflurane. The effects of the general anesthetics on PPF of PS amplitudes were examined using conventional methods (Nathan et al., 1990). Fig. 2 shows the amplitude ratios (PS2/PS1) as a function of interstimulus intervals in the absence (panel A) and presence (panel B) of BMI. Volatile anesthetics enhanced PPF mainly due to a decrease of PS1, whereas intravenous anesthetics reduced PPF. BMI completely abolished the effects of intravenous anesthetics on PPF, but only partially blocked the enhancement of PPF by volatile anesthetics.

206

K. Hirota et al. / Toxicology Letters 100–101 (1998) 203–207

hibited a use-dependent effect (desensitization) on PS amplitudes; these effects were not observed in the presence of BMI (data not shown). In contrast, sevoflurane did not exhibit use-dependent effects. Table 1 summarizes the results of the current study. All the anesthetics tested in the current study universally reduced PS amplitude, however differential actions of the volatile and intravenous anesthetics were demonstrated on the enhancement of PPF and use-dependent changes of PS amplitudes. The actions of the GABAA agonist, muscimol, were found to be identical to those of the volatile anesthetics (Table 1) except that the effects of the volatile anesthetics were not reversed with BMI.

4. Discussion

Fig. 3. The use-dependent changes in population spike (PS) amplitude by general anesthetics. The effects of thiopental (2 ×10 − 4 M) and sevoflurane (3.0 vol.%) on amplitudes were plotted as a function of time. Anesthetics were applied during the rest period (no stimulus). The stimulus (0.1 Hz) was applied 20 min later (indicated by arrows).

The differences between intravenous and volatile anesthetics on use-dependent changes of PS are shown in Fig. 3. Anesthetics were applied during the resting state (no electrical stimuli). Stimuli (0.1 Hz) were applied 20 min following administration of the anesthetic. Thiopental ex-

The present results clearly demonstrate differential actions between volatile and intravenous anesthetics on synaptic transmission in hippocampal CA1 pyramidal neurons. Since the effects of the intravenous anesthetics were antagonized in the presence of BMI, it is probable that GABAA receptors are involved in the primary actions of intravenous anesthetics. In contrast, since BMI failed to completely antagonize the effects of the volatile anesthetics, it is likely that other mechanisms in addition to GABAAergic modulation are involved. Similar responses between muscimol and the volatile anesthetics suggest that the actions of volatile anesthetics are due in part to enhancement of the postsynaptic GABAA-recep-

Table 1 Effects of general anesthetics on electrophysiologic properties in hippocampus Anesthetics

Concentration

PS amplitude

PS2/PS1 facilitation (PPF)

Use-dependency

Bicuculline blockade (GABAA)

Thiopental Pentobarbital Propofol

2×10−4 M 10−4 M 10−4 M

Decrease Decrease Decrease

Decrease Decrease Decrease

Yes Yes Yes

Yes Yes Yes

2.0 vol.% 2.0 vol.% 3.0 vol.% 2×10−5 M

Decrease Decrease Decrease Decrease

Increase Increase Increase Increase

No No No No

Partially Partially Partially Yes

Halothane Isoflurane Sevoflurane Muscimol (GABAA agonist)

K. Hirota et al. / Toxicology Letters 100–101 (1998) 203–207

tor as previously reported (Higashi and Nishi, 1982; Nakahiro et al., 1989). Stimulation using a paired-pulse paradigm reveals the facilitation in the second-evoked PS amplitude at the interstimulus interval of 40–200 ms. Albertson et al. (Albertson et al., 1991, 1992) reported depression of PPF of PS amplitudes by intravenous anesthetics in hippocampal preparations in vivo, and suggested that intravenous anesthetics enhance the GABAA-mediated inhibition. GABAA-mediated mechanisms, however, cannot explain the observed enhancement of PPF by the volatile anesthetics. Recent studies (Nathan et al., 1990; Davis et al., 1990; Nathan and Lambert, 1991) have demonstrated that PPF is due to the activation of GABAB receptors on presynaptic terminals, which results in reduction of release of the neurotransmitter (GABA) from these terminals. Therefore, volatile anesthetics may activate presynaptic GABAB receptors as well as GABAA receptors, and enhance PPF due to the GABAB-mediated feedback inhibition of GABA release. Activation of GABAB receptors by volatile anesthetics in hippocampal preparations has been demonstrated in other electrophysiologic studies (Pearce et al., 1989; Hirota and Roth, 1997a,b). Biochemical experiments have shown that many intravenous but not volatile anesthetics can inhibit the GABA uptake process in rat brain synaptosomes (Mantz et al., 1995). Intravenous anesthetic-induced depression of GABA uptake may increase GABA concentration in the synaptic cleft, and decrease the pool of GABA in inhibitory presynaptic terminals following stimulation. The elevated GABA concentration in the synaptic cleft might account for the significant depression of PPF by these agents. The depletion in the pool of GABA in presynaptic terminals might also explain the usedependent effects on PS amplitudes. In conclusion, the present results demonstrate that the actions of the volatile and intravenous anesthetics on evoked synaptic responses are different which may be a result of differential actions on presynaptic GABAergic mechanisms.

.

207

Acknowledgements We thank Zeneca, Cheshire, UK, for the gift of propofol, Dinabot, Osaka, Japan, for the gift of isoflurane and a isoflurane vaporizer, and Maruishi Pharmaceutical, Osaka, Japan for the gift of sevoflurane and a sevoflurane vaporizer. References Albertson, T.E., Tseng, C.C., Joy, R.M., 1991. Propofol modification of evoked/hippocampal dentate inhibition in urethane-anesthetized rats. Anesthesiology 75, 82 – 90. Albertson, T.E., Walby, W.F., Joy, R.M., 1992. Modification of GABA-mediated inhibition by various injectable anesthetics. Anesthesiology 77, 488 – 499. Davis, C.H., Davis, S.N., Collingridge, G.L., 1990. Paired-pulse depression of monosynaptic GABA-mediated inhibitory postsynaptic responses in rat hippocampus. J. Physiol. (Lond.) 424, 513 – 531. Higashi, H., Nishi, S., 1982. Effects of barbiturates on the GABA receptor of cat primary afferent neurons. J. Physiol. (Lond.) 332, 299 – 314. Hirota, K., Roth, S.H., 1997a. Sevoflurane modulates both GABAA and GABAB receptors in area CA1 of rat hippocampus. Br. J. Anaesth. 78, 60 – 65. Hirota, K., Roth, S.H., 1997b. The effects of sevoflurane on population spikes in CA1 and dentate gyrus of the rat hippocampus in vitro. Anesth. Analg. 85, 426 – 430. MacIver, M.B., Roth, S.H., 1988. Inhalation anaesthetics exhibit pathway-specific and differential actions on hippocampal synaptic responses in vitro. Br. J. Anaesth. 60, 680 – 691. MacIver, M.B., Kendig, J.J., 1991. Anesthetic effects on resting membrane potential are voltage-dependent and agent-specific. Anesthesiology 74, 83 – 88. Mantz, J., Lecharny, J.B., Laudenbach, V., Henzel, D., Peytavin, G., Desmonts, J.M., 1995. Anesthetics affect the uptake but not the depolarization-evoked release of GABA in rat striatal synaptosomes. Anesthesiology 82, 502 – 511. Nakahiro, M., Yeh, J.Z., Brunner, E., Narahashi, T., 1989. General anesthetics modulate GABA receptor channel complex in rat dorsal ganglion neurons. FASEB J. 3, 1850 – 1854. Nathan, T., Jensen, M.S., Lambert, J.D.C., 1990. GABAg receptors play a major role in paired-pulse facilitation in area CA1 of the rat hippocampus. Brain Res. 531, 55 – 65. Nathan, T., Lambert, J.D.C., 1991. Depression of the fast IPSP underlies paired-pulse facilitation in are CA 1 of the rat hippocampus. J. Neurophysiol. 66, 1704 – 1715. Pearce, R.A., Stringer, J.L., Lothman, E.W., 1989. Effects of volatile anesthetics on synaptic transmission in the rat hippocampus. Anesthesiology 71, 591 – 598.