The Actions of Sevoflurane and Desflurane on the γ-Aminobutyric Acid Receptor Type A: Effects of TM2 Mutations in the α and β Subunits

Mutagenesis of the GABA^Areceptor subunit cDNA was performed with the unique site elimination method as described previously. 17In brief, the unique site elimination method uses a two-primer system in which one primer is a mutagenic primer (Operon Technologies, Alameda, CA) and the other is a selection primer; a unique S sp  I site in the expression vector pCIS2 was mutated to an alternative restriction site (Mlu  I). These primers were 5′-phosphorylated using polynucleotide kinase, and then used to create mutations using the unique site elimination kit (Amersham Pharmacia Biotech, Piscataway, NJ). S sp  I digestion was used to select in favor of desired mutants and against template DNA; positive clones of transformed Escherichia coli  were screened for the appearance of the desired mutations by digestion with H pa  II. The sequences of all cDNA inserts were confirmed by double-stranded DNA sequencing.

Human embryonic kidney (HEK) 293 cells have been very useful for the transient expression and electrophysiologic analysis of GABA^Areceptors. 18,19The GABA^Areceptor subunit cDNAs were obtained as follows: human α1, human α2, rat β2, rat β3, and human γ2 from the laboratory of the late Dr. Dolan Pritchett. The wild-type or mutant GABA^Areceptor cDNAs were expressed via  the vector pCIS2, which contains one copy of the strong promoter from cytomegalovirus and a polyadenylation sequence from simian virus 40. These constructs were used to transfect HEK 293 cells (American Type Culture Collection, Rockville, MD). HEK 293 cells were maintained in culture on glass coverslips; cells were passaged weekly by trypsin treatment up to 20 times before being discarded and replaced with early passage cells. Cells was transfected using the CaPO⁴precipitation technique. 20Three to five micrograms of each cDNA was added to 65 μl of distilled water, 8 μl of 2.5 M CaCl²and 75 μl of 50 mm N,N-bis[2-hydroxyethyl]-2-aminoethanesulfonic acid for each coverslip. The precipitation reaction was allowed to proceed for approximately 60 min. The cDNA was in contact with the HEK cells for 24 h in an atmosphere containing 3% CO²(37°C) before being removed and replaced with fresh culture medium in an atmosphere of 5% CO²(37°C).

The coverslips were transferred, between 48 and 72 h after removal of the cDNA, to a large recording chamber and perfused continuously with extracellular solutions (145 mm NaCl, 3 mm KCl, 1.5 mm CaCl², 1 mm MgCl², 6 mm D-glucose, 10 mm HEPES/NaOH adjusted to pH 7.4). Whole cell patch clamp recordings from HEK 293 cells were made using the Axopatch 200 amplifier (Axon Instruments, Foster City, CA) as described previously. 7The resistance of the patch pipette was 4–6 MΩ when filled with internal solutions (145 mm N -methyl-d-glucamine HCl, 5 mm dipotassium ATP, 1.1 mm EGTA, 2 mm MgCl², 5 mm HEPES/KOH, 0.1 mm CaCl²adjusted to pH 7.2). Cells were voltage clamped at −60 mV. Because the chloride concentrations in the intracellular and extracellular solutions were almost identical, the calculated chloride equilibrium potential was approximately 0 mV. In addition to the continuous bath perfusion with extracellular medium, solutions including GABA and/or volatile anesthetics were applied to the cell by local perfusion using a motor-driven solution-exchange device (Rapid Solution Changer RSC-100; Molecular Kinetics, Pullman, WA). This system exchanges solutions around the cells within approximately 20–50 ms. 21Laminar flow out of the rapid solution changer head was achieved by driving all solutions at identical flow rates (1–1.5 ml/min) via  a multichannel infusion pump (Stoelting, Wood Dale, IL). The solution changer was driven by protocols in the acquisition program of pCLAMP version 5 (Axon Instruments). For GABA concentration-response studies, cells were typically superfused with extracellular saline before switching into one of seven GABA concentrations for 3 s, and returning to saline for at least 15 s before any subsequent GABA application. GABA responses typically exhibited a small degree of desensitization during agonist applications, such that the amplitude of the responses declined by 10–15% from the peak current. However, recovery from desensitization was typically rapid after agonist removal, and GABA current amplitudes were stable in response to repeated applications. Currents were low-pass filtered at 2 kHz and digitized with the TL-1–125 interface (Axon Instruments) using pCLAMP 5 and stored for off-line analysis. All experiments were performed at room temperature (21°–24°C).

Stock solutions of GABA and anesthetic compounds were diluted into the extracellular solutions daily shortly before use. All volatile anesthetic solutions were prepared by injection of liquid anesthetic with a gas-tight syringe (Hamilton, Reno, NV) into intravenous drip bags containing defined volumes of extracellular solutions (100 ml) and used for up to 4 h. 21The clinically relevant “minimum alveolar concentration (MAC) equivalent” aqueous concentrations defined for this study and most often applied to cells were isoflurane 370 μm, sevoflurane 360 μm, and desflurane 280 μm. Losses of volatile anesthetics in this perfusion system has been measured using gas chromatography and represents only 5–10% of the total applied drug concentration. 19The sources of the anesthetics used were as follows: Ayerst Laboratories (Philadelphia, PA) for halothane, Abbott Laboratories (North Chicago, IL) for isoflurane and sevoflurane, Ohmeda (Liberty Corner, NJ) for desflurane.

Control GABA concentration-response data were expressed as a fraction of the maximal response to GABA in each cell, allowing normalized data from different cells to be combined. Pooled data were fitted using a weighted sum of least-squares method (KaleidaGraph version 3.5, Synergy Software, Reading, PA) to a Hill equation:

where I / I  max is the whole cell current amplitude expressed as a percentage of the current maximal peak current, [drug] is the GABA concentration, EC⁵⁰is the GABA concentration eliciting a current equal to half amplitude of I  max, and nH is the Hill coefficient. Anesthetic-induced potentiation of a submaximal GABA response was then calculated as the percentage increase above the control (EC²⁰) response to GABA in the presence of the anesthetics. Anesthetics were always preapplied for 3 s before coapplication with agonist to ensure that the anesthetic had reached equilibrium with the receptors. All data are presented as mean ± SEM with n  number of cells tested. Statistical significance was determined using one-way ANOVA to assess differences between groups.

The wild-type GABA^Aα1β2γ2s receptor, α2β3γ2s receptor and receptors harboring mutations in TM2 segment such as α1(S270W)β2γ2s, α1β2(N265W)γ2s and α2(S270I)β3γ2s were transiently expressed in HEK 293 cells. Large inward chloride currents induced by GABA (0.01 μm to 1 mm) were recorded in the absence of the anesthetics, at a holding potential of −60 mV. Representative recordings of GABA-induced Cl⁻currents are presented for the wild-type α1β2γ2s receptor (fig. 2A, top) and for the wild-type α2β3γ2s receptor (fig. 2A, bottom). To examine the effects of mutation in TM2 of the α or β subunit on the apparent affinity for the agonist, full concentration-response curves for GABA were determined in all receptors. Currents elicited by at least seven different concentrations of GABA were expressed as a fraction of the maximal GABA response, and the normalized data were fitted by a Hill equation. Figures 2B and Cdisplay GABA concentration-response curves for the wild-type receptor and for receptors mutated at TM2 of either α or β subunit. Mutation of Ser270 to a large amino acid such as tryptophan or isoleucine produced a leftward shift in the curves, consistent with an increased apparent affinity for GABA. The GABA EC⁵⁰, Hill coefficient and maximal current amplitude data for all receptors tested are summarized in table 1. There was no significant difference in Hill coefficient or in the amplitude of peak GABA currents between the wild-type receptor and mutant receptors.

To determine whether the mutated amino acid residues influenced the ability of volatile anesthetics to potentiate GABA^Areceptor function, the effects of desflurane, sevoflurane, and isoflurane on the amplitude of GABA responses were examined. Figure 3Ashows representative recordings from experiments in which we measured the potentiation of submaximal (EC²⁰concentration) GABA-induced currents by sevoflurane and desflurane in cells expressing the wild-type α1β2γ2s receptor. Clinically relevant concentrations of isoflurane (1 MAC ≅ 370 μm), sevoflurane (1 MAC ≅ 360 μm), and desflurane (1 MAC ≅ 280 μm) significantly enhanced the amplitude of GABA responses by 99.0 ± 13.1, 59.3 ± 11.2, and 55.7 ± 12.6% above control, respectively (fig. 3B). These data are consistent with our previous observations that clinical concentration equivalents of isoflurane potentiated EC²⁰GABA currents by approximately 100%. 7When Ser270 of the α1 subunit was substituted by a large amino acid tryptophan (W), the mutant receptor α1(S270W)β2γ2s showed no significant potentiation by isoflurane, sevoflurane, and desflurane (8.2 ± 7.1, 7.5 ± 5.5, 3.6 ± 9.8% above control, fig. 3B, bottom), suggesting that α1(S270) is one of the key residues for the modulation of GABA^Areceptors by these volatile anesthetics. In fact, no significant potentiation was observed in the α1(S270W)β2γ2s receptor for any of the three volatile agents, up to 2 mm anesthetic.

To confirm the importance of Ser270 of the α subunit, we also studied the effects of this mutation in another subunit combination. The α2β3γ2s receptor subtype was used because this is the second most abundant subunit combination in mammalian brain. 11As shown in figure 2Cand table 1, the α2β3γ2s GABA receptor also expressed well in HEK cells. As shown in figure 4A, the three volatile anesthetics significantly enhanced GABA-induced currents in the α2β3γ2s wild-type receptor (89.2 ± 15.1, 49.3 ± 15.5, and 65.3 ± 22.1% of control, n ≥ 8, P < 0.001 for each anesthetic). The degree of potentiation was similar to that observed in α1β2γ2s receptor (fig. 3). When Ser270 of the α2 subunit was substituted by the large amino acid isoleucine (I), the mutant receptor α2(S270I)β3γ2s was insensitive to potentiation of GABA responses by sevoflurane and desflurane (7.6 ± 8.8, 3.2 ± 7.6, and 5.5 ± 4.3% above control, respectively, fig. 4B).

As shown in figure 1, Asn265 in the β2 subunit is homologous with position Ser270 in the α1 subunit. In fact, this position is an important amino acid residue for intravenous anesthetic modulation of the receptor. 12,13,22We therefore examined the effects of TM2 mutation in the β2 subunit using α1β2(N265W)γ2s. In contrast to α1(S270W)β2γ2s, isoflurane, sevoflurane, and desflurane potentiated GABA-induced currents strongly in the α1β2(N265W)γ2s receptor (fig. 5), although the degree of potentiation produced by these anesthetics was somewhat smaller than that measured for the wild-type α1β2γ2s receptor.

All GABA^Areceptors tested in this study were successfully expressed in HEK 293 cells and the apparent affinity for GABA could be measured easily. We believe from our data that amino acid substitutions in the TM2 segment of the α1, α2 or β2 subunits of the GABA^Areceptor did not compromise ion channel function significantly, consistent with the notion that these point mutations did not produce large structural changes in the region of the channel pore, or in the GABA binding sites of the extracellular N-terminus.

As illustrated in figure 3A, clinically relevant concentrations of isoflurane, sevoflurane, and desflurane significantly increased the amplitude of submaximal GABA-induced currents in the wild-type GABA^Aα1β2γ2s receptors. These data are consistent with previous observations that volatile anesthetics potentiated GABA^Areceptor-mediated chloride currents in recombinant GABA^Aα2β1 receptor, 4,23GABA^Aα1β2γ2s receptor, 7,24GABA^Areceptors of cultured neurons, 25,26and in brain slices. 27,28These data are consistent with the hypothesis that neuronal inhibition mediated by GABA^Areceptors plays an important role in the production of anesthesia by these agents.

When the GABA^Areceptor α1(S270W) β2γ2s was tested for modulation by sevoflurane and desflurane, there was a strong effect of the larger side chains such as tryptophan, which completely abolished the anesthetic potentiation by sevoflurane and desflurane (fig. 3A, bottom). These data confirm previous findings that mutation of α2(S270) in α2β1 can ablate potentiation by the inhaled anesthetic enflurane, 3isoflurane, 4,23and halothane, 4suggesting that S270 of the α subunit is a key residue for potentiation of GABA^Areceptors by sevoflurane and desflurane, and for other volatile anesthetics. In contrast, GABA-induced currents in a receptor mutated in the TM2 segment of the β subunit, α1β2(N265W)γ2s, retained sensitivity to the anesthetics (fig. 5). The contrasting effects of α1(S270) and β2(N265) subunit mutations in the current study suggest that it is more likely that N265 governs critical allosteric transitions, rather than directly binding volatile anesthetics.

The importance of Ser270 of the α subunit was confirmed not only in the α1β2γ2s receptor, but also in the α2β3γ2s receptor. The mutations analyzed in the current study all result from the replacement of a smaller amino acid by a larger amino acid (e.g. , serine to isoleucine, serine to tryptophan, asparagine to tryptophan). If residues within TM2 of the GABA α subunit do indeed form part of a binding pocket, then substitution by larger amino acid residues at these positions should hinder volatile anesthetic binding, and thereby ablate positive modulation of receptor function. We therefore hypothesize that the serine residue in TM2 of α1 and α2 subunits forms part of a binding site not only for halothane and isoflurane, but also for sevoflurane and desflurane. The molecular volumes for desflurane and sevoflurane are very similar to that of isoflurane, so that the isoflurane-like pattern of activity of these drugs in the wild-type and mutant receptors studied here is not unexpected. Further investigation will be needed verify whether general anesthetics such as sevoflurane and desflurane do indeed bind directly to amino acids in the TM2 segment of the α subunit. A high-resolution structure of the GABA^Areceptor must first be resolved.

Site-directed mutagenesis studies have demonstrated the importance of the residue N265 in the β subunit for potentiation of GABA responses by etomidate and the anticonvulsant loreclezole 22,29. In fact, recent work has demonstrated that mutation of Asn289 within the GABA^Aβ3 subunit (homologous with Asn265 in the β2 subunit) to methionine abolishes direct activation and GABA-induced potentiation by etomidate in vitro , 12but the involvement of the β subunit in regulation by volatile anesthetic is still unclear at this time. 7In addition, a knock-in mouse harboring the N289 M mutation in the β3 subunit shows a drastically reduced behavioral response to both propofol and etomidate, 13using both the loss of righting reflex and withdrawal reflex assays. The evidence for involvement of the GABA^Areceptor in these behavioral actions of etomidate and propofol therefore appears incontrovertible, and this strengthens the case in favor of similar knock-in mouse experiments with the volatile anesthetics studied here.

In conclusion, the data presented here confirm previous reports that Ser270 of the α1 subunit is important in the actions of inhaled agents at the GABA^Areceptor, and extends these observations to sevoflurane and desflurane. In addition, we provide new information that the importance of Ser270 of the α subunit is conserved in the wild-type GABA^Aα2β3γ2s receptor. Several lines of evidence have converged to infer the existence of a common site of action for halothane and isoflurane within the α subunit. 4,7,24,30,31The present data are consistent with the existence of a cavity that may be occupied by isoflurane, sevoflurane, or desflurane because the same mutations block potentiation by all three agents.

The authors thank Alyson Andreasen, B.S., Nicole V. Baron, B.S., and Gina E. Radoi, B.S. (Technicians, Weill Medical College of Cornell University) for technical assistance and Hiroko Suzuki, B.A. (Secretary, Gunma University School of Medicine) for secretarial assistance.