Despite significant recent advances, the mechanisms of volatile anesthetic action are still poorly understood (Hemmings et al., 2005). Recent evidence implicates mammalian voltage-gated Na⁺ channels (Na_v) as targets for the presynaptic depression of neurotransmitter release by volatile anesthetics (Ouyang et al., 2003, 2005; Wu et al., 2004; Hemmings et al., 2005). Voltage-gated Na⁺ channels are critical for the initiation and conduction of action potentials in excitable cells (Hodgkin and Huxley, 1952; Hille, 2001), and they are important in regulating presynaptic excitability (Meir et al., 1999) and neurotransmitter release (Tibbs et al., 1989). Nine homologous subtypes of the four-domain pore-forming Na_v α subunit have been identified in mammals (Yu and Catterall, 2003). A single-domain prokaryotic voltage-gated Na⁺ channel (NaChBac) with similar properties to mammalian voltage-gated ion channels has been isolated from Bacillus halodurans (Ren et al., 2001). The single domain of the presumably tetrameric NaChBac consists of six transmembrane α-helical segments structurally homologous to the six transmembrane segments of each of the four domains of voltage-gated Na⁺ and Ca²⁺ channels.

NaChBac exhibits the basic kinetic features of Na_v: it opens with depolarization in a voltage-dependent manner (activation) over a comparable voltage range; closes upon prolonged depolarization (inactivation); and returns to a resting, nonconducting state with repolarization (deactivation) (Ren et al., 2001). However, the kinetic behavior of NaChBac is slower than that of eukaryotic Na_v channels, including ∼6-fold slower activation, ∼17-fold slower inactivation, and ∼10-fold slower recovery from inactivation (Ren et al., 2001; Kuzmenkin et al., 2004). These conserved basic gating mechanisms with slower kinetics provide a potential advantage in studying NaChBac as a model for Na_v because of the intrinsic demands of doing experiments with channels that have substantially faster gating.

NaChBac activation is preceded by movement of a gating charge, an essential feature of voltage-gated ion channels (Kuzmenkin et al., 2004). The relatively simple amino acid sequence of NaChBac (274 amino acids; 31 kDa), its structural homology to Na_v, and the ease of producing large amounts in bacterial culture provide a potential opportunity for structure-function studies and three-dimensional structure determinations by X-ray crystallography of a model anesthetic target. This led us to investigate the effects of isoflurane, a prototypical inhaled anesthetic ether, on the basic functional properties of heterologously expressed NaChBac. Isoflurane is representative of the modern family of halogenated ether anesthetics currently in widespread clinical use that are thought to share their essential molecular targets for the production of general anesthesia (Hemmings et al., 2005).

Materials and Methods

Heterologous Expression of NaChBac. The cDNA for wild-type NaChBac in a modified pTracer expression vector (6.2 kilobases, containing green fluorescent protein-Zeocin site; Invitrogen, Carlsbad, CA), generously provided by Dr. D. Clapham (Harvard University, Boston, MA), was amplified in bacterial culture, and then the cDNA was purified. HEK 293 cells (American Type Culture Collection, Manassas, VA) were transiently transfected using Lipofectamine 2000 (Invitrogen) and cultured in Dulbecco's modified Eagle's medium/F-12 (Invitrogen) with 5% (v/v) fetal bovine serum (BioSource International, Camarillo, CA) and 5% CO₂, 95% O₂ (v/v) at 37°C. One to 3 days before electrophysiological recording, cells were plated on glass coverslips in 35-mm plastic dishes (BD Biosciences, Franklin Lakes, NJ). Transfected cells were identified using a Nikon ECLIPSE TE300 inverted fluorescence microscope (Nikon, Melville, NY).

Electrophysiological Recordings. The culture medium was changed, and cells were superfused at 2 to 3 ml/min with extracellular solution containing 140 mM NaCl, 4 mM KCl, 1.5 mM CaCl₂, 1.5 mM MgCl₂, 10 mM HEPES, and 5 mM d-glucose, pH 7.30 with NaOH. Na⁺ currents (I_NaChBac) were recorded at room temperature (24 –26°C) using the whole-cell patch-clamp technique (Hamill et al., 1981). Patch electrodes were made from borosilicate glass capillaries (Drummond Scientific, Broomall, PA) using a four-step protocol (P-97 micropipette puller; Sutter Instrument Company, Novato, CA), fire polished (Narishige Microforge, Kyoto, Japan) and coated with SYLGARD (Dow Corning, Midland, MI) to reduce background noise and capacitance. The pipette electrode solution contained 80 mM CsF, 40 mM CsCl, 15 mM NaCl, 10 mM HEPES, and 10 mM EGTA, pH 7.35 with CsOH. Cells <15 μm in diameter were selected for recording using low resistance electrodes (1–3 MΩ). To determine the contribution of steady-state inactivation to isoflurane inhibition I_NaChBac, two holding potentials (–80 or –100 mV) were used. Because of the slow current decay observed with consecutive pulses, low-frequency stimuli (0.2 or 2 Hz) were used to study use-dependent block of I_Na. Currents were sampled at 10 or 20 kHz and filtered at 1 to 3 kHz using an Axon 200B amplifier, digitized with a Digidata 1321A interface, and analyzed using pClamp 8.2 software (Molecular Devices, Sunnyvale, CA). Capacitance and 60 and 85% series resistance were compensated, and leak current was subtracted using P/4 or P/5 protocols. Cells were held at –80 mV between recordings with pulse frequencies of 0.1 to 2 Hz.

Isoflurane (Abbott Laboratories, North Chicago, IL) was diluted from stock solutions in 10 to 12 mM extracellular solution (prepared by stirring at room temperature for 12–24 h) into airtight glass syringes. It was applied locally to attached cells at 0.05 ml/min through a 0.15-mm-diameter perfusion pipette located 30 to 40 μm from patched cells using an ALA-VM8 pressurized perfusion system (ALA Scientific Instruments, Westbury, NY). Isoflurane concentrations were determined by local sampling of the perfusate at the site of the recording pipette tip and by analysis by gas chromatography following n-heptane extraction; measured isoflurane concentrations were ∼85% of syringe concentrations.

Data Analysis. IC₅₀ values (± S.E.M.) for isoflurane were determined by least-squares fitting of data to the Hill equation: I_Na/I_{Na Control} = 1/(1 + 10^((log IC₅₀ – X) × n_H)), where X is isoflurane concentration, I_Na/I_{Na Control} is normalized Na⁺ current, and n_H is Hill slope. Normalized activation data were fitted to a Boltzmann equation: G/G_max = 1/(1 + exp(V_1/2a – V)/k), where G/G_max is normalized fractional conductance, G_max is maximal conductance, V_1/2a is voltage for half-maximal activation, and k is slope factor. Na⁺ conductance (G_Na) was calculated as G_Na = I_Na/(V_t – V_r), where I_Na is peak Na⁺ current, V_t is test potential, and V_r is Na⁺ reversal potential (E_Na = 67 mV). Inactivation curves were fitted to a Boltzmann equation: I_Na/I_Na max = 1/(1 + exp(V – V_1/2in)/k), where I_Na/I_{Na max} is normalized current, I_{Na max} is maximal current, V_1/2in is voltage of half-maximal inactivation, V is test potential, and k is slope factor. I_Na inactivation was analyzed by fitting to a monoexponential equation: I_Na = A·exp(–t/τ) + C, where A is maximal I_Na amplitude, C is plateau I_Na, t is time, and τ is the time constant of current decay. NaChBac current traces were also fitted to the conventional Hodgkin and Huxley m³h model (Hodgkin and Huxley 1952) in modified form: I(t) = I_o + I_max × (1 – exp(–(t – t_o)/τ))³_act × exp(–(t – t_o)/τ_inact), where I(t) is I_Na as a function of time, I_o is I_Na at t = t_o, I_max is maximal current, t_o is initial time of I_Na rising phase, t is time during the pulse, and τ_act and τ_inact are activation and inactivation time constants, respectively. The time course of use-dependent decay of I_Na was analyzed by fitting to a monoexponential equation: Normalized I_Na = exp(–time constant × t) + C, t is pulse number, C is plateau I_Na, and time constant is the time constant of use dependent of decay. Data were analyzed using pCLAMP 8.2, Prism version 4.0 (GraphPad Software Inc., San Diego, CA), SigmaPlot version 9.0 (Systat Software, Inc., San Jose, CA), and Origin 7.5 SR4 (OriginLab Corp., Northampton, MA). Statistical significance (p < 0.05) was assessed by analysis of variance or paired t test, as indicated.

Results

Basic Properties of NaChBac. HEK 293 cells expressing NaChBac showed voltage-gated I_Na up to 10 nA (Fig. 1). Cells with endogenous Na⁺ currents (amplitude <0.25 ± 0.08 nA; n = 8), easily identified by their faster activation compared with I_Na, were discarded. To avoid large series resistance and space-clamp error, only cells with I_Na of 0.5 to 5 nA were analyzed. I_Na activated at –60 mV or higher. Peak activation was at 0 mV from a holding potential of –80 mV, and at –10 mV from a holding potential of –100 mV, where a large fraction of channels are available and in the resting state (Ren et al., 2001; Fig. 1B). Peak I_Na occurred 10 to 20 ms from the beginning of the depolarizing pulse, which is consistent with a report that NaChBac activation is severalfold slower compared with Na_v (Ren et al., 2001). I_Na reversed at +70 mV (Fig. 1B), near the calculated Nernst equilibrium potential for Na⁺ for the conditions used (+67 mV). I_Na inactivation developed over 700 to 850 ms (Fig. 2).

Isoflurane Inhibits NaChBac. Isoflurane reversibly inhibited I_Na with no change in the voltage of peak activation or reversal potential (Figs. 1 and 2). The effects of isoflurane on I_Na were similar from a holding potential of –80 or –100 mV. The concentration-effect curves were well fitted by the Hill equation (Fig. 2B), and they yielded IC₅₀ values near the clinically effective concentration of isoflurane (aqueous concentration equivalent to minimal alveolar concentration = 0.35 mM in rat; Taheri et al., 1991): 0.35 ± 0.03 mM from a holding potential of –80 mV and 0.48 ± 0.02 mM from a holding potential of –100 mV (p < 0.01; n = 4–12).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Effects of isoflurane on NaChBac current-voltage relationships. A, isoflurane (∼0.8 mM; equivalent to ∼2.3 times minimal alveolar concentration in rat) significantly inhibited I_Na from a holding potential of either –80 or –100 mV. Effects were analyzed at various phases of the test pulse: 5 ms after depolarization (A), peak I_Na at 10 to 20 ms (B), end I_Na 29 ms after depolarization (C), and tail current at repolarization (D). B, voltage for peak inward current (V_max) was 0 mV from a holding potential of –80 mV (left) and –10 mV from a holding potential of –100 mV (right). I_Na was activated by 30-ms test pulses of –60 to +70 mV in 10-mV steps. Mean isoflurane concentrations in B were 0.79 ± 0.06 mM for holding potential of –80 mV and 0.82 ± 0.05 mM for holding potential of –100 mV (mean ± S.E.M.; n = 7–8).

Onset of inhibition of I_Na by local pipette application of isoflurane was rapid (<1 min) and reversed within minutes of washout (Fig. 3). Inactivation of I_Na was highly sensitive to isoflurane (Figs. 2A and 3). Inactivation was best fitted by a monoexponential function with a time constant (τ) that was markedly reduced by isoflurane, consistent with facilitation of the inactivation process (see below). Isoflurane binding affinity during channel inactivation (K_D = 0.06 mM; Fig. 2C) was 8-fold higher than the IC₅₀ for peak current amplitude (IC₅₀ = 0.48 ± 0.02 mM, from a holding of –100 mV; Fig. 2B), which reflects binding affinity to open channels. The marked reversible increase in current decay induced by isoflurane was also observed with repetitive stimulation (Fig. 3C). The linear relationship between blocking rate (1/τ) and isoflurane concentration (Fig. 2C) reflects a bimolecular interaction between isoflurane and its binding site (Ramos and O'Leary, 2004).

Effects of Isoflurane on Channel Gating. The effects of isoflurane on NaChBac channel gating kinetics were analyzed from the calculated conductance and conductance ratios at each of two points (peak and end) during a 30-ms depolarizing pulse (Fig. 4). Conductance and normalized conductance activation data were best fitted by a simple Boltzmann function. Inhibition at peak current and at the end of the depolarizing pulse more accurately reflect channel kinetic changes induced by isoflurane (see Figs. 1 and 4A). Isoflurane (∼0.8 mM) produced a depolarizing shift in the voltage dependence of peak I_Na activation and a hyperpolarizing shift in the voltage dependence of end I_Na activation (Fig. 4B; Table 1). The shift in voltage of end steady-state current activation (V_1/2aend) by isoflurane was greater from a holding potential of –100 mV and of voltage of half-maximal peak current (V_1/2apeak) from –80 mV (Table 1). This could be due to a greater fraction of inactivated channels at a holding potential of –80 mV (Fig. 5) and to more newly inactivated channels at the end of the pulse at a holding potential of –100 mV, respectively. The net effect of isoflurane was to narrow the current window of activation for either holding potential, thus decreasing the peak current.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Isoflurane reversibly inhibits NaChBac at clinically relevant concentrations. A, representative current traces showing reversible inhibition by isoflurane of I_Na from a holding potential of –100 mV. B, concentration-effect curves for inhibition of normalized peak I_Na by isoflurane from holding potentials of –80 or –100 mV. Data (mean ± S.E.M.; n = 4–12) were fitted to the Hill equation. **, p < 0.01 by Student unpaired t test. C, isoflurane-facilitated current decay was fitted to a monoexponential equation (see Materials and Methods). The blocking rate (1/τ · 1000) plotted against isoflurane concentration was fitted to a linear equation to yield the isoflurane binding affinity (K_D = k_off/k_on = 0.06 mM); the slope and y-intercept yielded the association rate (k_on = 98.4 s^–1 mM^–1) and dissociation rate (k_off = 5.5 s^–1), respectively. Currents were evoked by 850 ms test pulses to V_max (–20 to 0 mV).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Reversible inhibition of NaChBac by isoflurane with repetitive stimulation. I_Na was activated from a holding potential of –100 mV by 850-ms test pulses to –10 mV at 8.5-s intervals between sweeps. A, representative traces showing time course of I_Na inhibition over 106 pulses. B, normalized peak I_Na amplitudes showing reversible inhibition of I_Na by 0.41 mM isoflurane. The black line indicates the period of isoflurane application (1.5 min). C, reduction and recovery of time constant of current inactivation (τ) by isoflurane.

Isoflurane reduced channel availability as reflected in the marked hyperpolarizing shift in the voltage dependence of inactivation from a holding potential of –80 mV with no change in slope factor (Fig. 5; Table 1). With a 2-s prepulse from –80 mV, ∼26% of channels were inactivated, consistent with greater isoflurane effect early (5 ms; data not shown) in the depolarizing pulse. From a holding potential of –100 mV, greater block of I_Na by isoflurane occurred later in the depolarizing pulse (end of pulse at 29 ms; Table 1) due to greater current decay from enhanced channel inactivation.

The effects of isoflurane on NaChBac channel gating were also modeled by fitting to Hodgkin and Huxley m³h kinetics. This approach provides only a phenomenological description, but it is useful in evaluating relative effects on activation and inactivation kinetics. Increasing isoflurane shifted the time to peak I_Na to earlier times (Fig. 6A, inset, circles), revealing an accelerating effect on current time course. The time to peak I_Na was reduced ∼25% by 0.83 mM isoflurane (n = 2). From the fit of the complete current time course, τ_act/τ_ctl decreased ∼30%, whereas τ_inact/τ_ctl decreased ∼95% for 0.83 mM isoflurane (n = 2; Fig. 6B). Both effects contributed to the reduction of peak I_Na by isoflurane (Fig. 6B, inset). The time to half activation was less sensitive to isoflurane than the time to peak I_Na (Fig. 6A, inset). This suggests rapid tonic block of open channels during activation before current reaches peak amplitude and/or tonic block of closed channels, which cannot be distinguished by these data.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Effects of isoflurane on NaChBac activation. A, NaChBac conductance was calculated (see Materials and Methods) and plotted for two time points during the depolarizing pulse as shown in Fig. 1A. Peak at 10 to 20 ms and end of pulse at 29 ms. Conductance was plotted versus the 30-ms depolarizing test voltage (–60 to +40 mV). B, conductance was normalized to its maximal control value and fitted to a Boltzmann equation. Mean isoflurane concentrations were 0.79 ± 0.06 mM for holding potential of –80 mV and 0.82 ± 0.05 mM for holding potential of –100 mV (mean ± S.E.M.; n = 7–8).

Use-Dependent Block of NaChBac by Isoflurane. NaChBac exhibited greater use-dependent current decay at 2-versus 0.2-Hz stimulus frequency (Table 2, upper), which reflects the slow recovery of NaChBac from inactivation. Isoflurane enhanced the use-dependent decay of I_Na and increased the time constant (Table 2, lower). Isoflurane also produced tonic block evident in reduced normalized I_Na amplitudes (Fig. 7; Table 2). The enhanced use-dependent reduction of residual current, expressed as normalized I_Na plateau, by isoflurane was greater from a holding potential of –80 mV (ΔC =–0.31) than from –100 mV (ΔC =–0.14; Table 2), consistent with enhanced channel inactivation and slower recovery from inactivation at a holding potential of –80 mV.

Discussion

Volatile anesthetics inhibit native neuronal and recombinant mammalian voltage-gated Na⁺ channels in a voltage-dependent manner by enhanced inactivation and tonic block (Rehberg et al., 1996; Ratnakumari et al., 2000; Ouyang et al., 2003; Shiraishi and Harris, 2004; Ouyang and Hemmings, 2005). We now show that isoflurane inhibits the prokaryotic voltage-gated Na⁺ channel NaChBac at concentrations that are comparable with those that block neuronal (Ratnakumari et al., 2000; Ouyang et al., 2003; Ouyang and Hemmings, 2005) and heterologously expressed mammalian Na⁺ channels (Rehberg et al., 1996; Stadnicka et al., 1999; Shiraishi and Harris, 2004). The simple single-domain structure, homologous pore region, and slower gating kinetics compared with eukaryotic channels make NaChBac an excellent model for structure-function studies of the pharmacology and ion channel binding sites of volatile anesthetics. The similar pharmacological sensitivity to isoflurane suggests that the site(s) of volatile anesthetic interaction with voltage-gated channels is conserved from bacteria to mammals.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

Effect of isoflurane on NaChBac inactivation. A, representative curves for steady-state inactivation before and after application of 0.78 mM isoflurane. Inactivation was measured using a two-pulse protocol in which cells were held at –80 mV and exposed to a 2-s conditioning pulse from –100 to 0 mV, followed by a 30-ms test pulse to –20 or to –10 mV to determine channel availability. B, normalized I_Na was fit to a Boltzmann equation to yield voltage of 50% maximal inactivation (V_1/2in). Isoflurane significantly shifted the voltage dependence of steady-state inactivation in the negative direction, with no effect on slope factors. Mean isoflurane concentration was 0.83 ± 0.06 mM. Data expressed as mean ± S.E.M. (n = 5). ***, p < 0.001 by paired t test.

Isoflurane and other volatile anesthetics affect multiple molecular targets critical to neuronal signaling in addition to Na⁺ channels, in particular GABA_A and glycine receptors, the principal inhibitory ligand-gated ion channels (Hemmings et al., 2005). Resolution of the relative contributions of various anesthetic-sensitive targets to the spectrum of pharmacological effects has been greatly facilitated by the design of anesthetic-insensitive mice harboring site-specific mutations in GABA_A receptors identified by structure-function studies (Jurd et al., 2003). This approach in the much larger four-domain voltage-gated ion channels will be greatly facilitated by models such as NaChBac. Although NaChBac differs in some respects from eukaryotic Na⁺ channels, elucidation of anesthetic mechanisms involving eukaryotic channels will be facilitated by looking at a simpler channel. The observations that NaChBac is inhibited by isoflurane at similar concentrations as Na_v and that the mechanisms are similar to those observed for Na_v validates this approach. Because NaChBac is selective for Na⁺, it is reasonable to assume that features of the pore structure will be similar between NaChBac and other Na⁺ channels. Similarities and differences between the mechanisms of block of NaChBac relative to other Na⁺ channel blockers will be informative in understanding the pharmacology of both channel types. Because NaChBac is presumed to be a homotetramer (by analogy with K⁺ channels), it is easier to mutate regions of interest for structure-function studies in this smaller protein than it is with the eukaryotic Na⁺ channels (Zhao et al., 2004). NaChBac can also be expressed in bacteria; hence, large amounts of the protein can be produced that will facilitate finer structural studies. NaChBac expresses well in mammalian cells so mutants can be easily tested.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

Differential effect of isoflurane on NaChBac activation and inactivation kinetics. A, representative currents elicited by pulses to –20 mV from a holding potential of –100 mV. Superimposed on each trace (in gray) is the fit of the current data to I(t) = I_o + I_max·(1 – exp(–(t – t_o)/τ_act))³·exp(–(t – t_o)/τ_inact) (see Materials and Methods). Inset, time to half activation (I_midpt; ▵, ▴) and the time to I_peak (○, •) as a function of isoflurane concentration. Data from two separate experiments are indicated by open versus closed symbols. τ_Imidpt and τ_Ipeak returned to control values after wash (data not shown). B, activation and inactivation time constants (τ_act and τ_inact, respectively) relative to control (ctl) plotted versus isoflurane concentration. Data from each of two independent experiments are indicated by open versus closed symbols. Inset, fractional maximal current amplitude (I/I_ctl) versus isoflurane concentration fitted by linear regression to y =–0.40x + 0.99 (set 1, •) or y =–0.52x + 1.04 (set 2, ▵).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7.

Use-dependent block of NaChBac by isoflurane. Isoflurane increased use-dependent decay of I_Na more from a holding potential of –80 mV (left) than from –100 mV (right). Graphs show that peak amplitude of each pulse normalized to that of the first pulse (mean ± S.E.M.; n = 6–11), plotted against stimulus number, and fitted to a monoexponential function (see Materials and Methods). Protocol: 30 ms, 2-Hz test pulses from holding potential (–80 or –100 mV) to peak activation voltage (–20 to 0 mV) in 10-mV steps.

The effect of isoflurane on NaChBac activation differs from its effects on mammalian Na_v (Rehberg et al., 1996; Stadnicka et al., 1999; Ouyang et al., 2003; Shiraishi and Harris, 2004), which exhibit no change in channel activation kinetics or activation curves. The rising and falling faces of NaChBac currents could be reproduced reasonably well using a Hodgkin-Huxley m³h kinetic model for classical nerve type Na⁺ channels (Hodgkin and Huxley 1952), even though some nerve-type Na⁺ channels are not fit well by the m³h kinetic model (Herzog et al., 2001). Although m³h kinetics do not accurately describe the activation time course because of transitions between multiple closed states before channel opening in neuronal Na⁺ channels (Baranauskas and Martina, 2006) that are also present in NaChBac (Kuzmenkin et al., 2004), the m³h model allows a simple phenomenological comparison of relative effects on activation and inactivation. Although a change in activation kinetics was apparent, it was not as large as the speeding of inactivation. However, a significant shift in the voltage dependence of activation and of inactivation was observed. Because NaChBac activates ∼6-fold slower than mammalian Na_v (Ren et al., 2001), the mechanism of tonic block of NaChBac by isoflurane might differ from its effects on mammalian Na_v by a greater contribution of open channel block during activation. Analysis of the mechanisms underlying isoflurane effects on NaChBac activation will require detailed gating current and single channel studies.

Isoflurane dissociates quickly from mammalian Na⁺ channels, which require rapid repetitive depolarization to develop use-dependent block (unpublished observations). The slower inactivation of NaChBac compared with Na_v resulted in enhanced use-dependent block by isoflurane at relatively low-stimulus frequency. These results are consistent with the modulated receptor hypothesis of Na⁺ channel block (Hille, 1977; Hondeghem and Katzung, 1977) in which the inactivated state is selectively stabilized by isoflurane. Open channel block by isoflurane is also probably due to the slower activation and inactivation kinetics of NaChBac. The NaChBac tail current was inhibited by isoflurane, and the tail conductance ratio overlapped with the pseudo steady-state conductance ratio, suggesting that enhancement of inactivation is the predominant mechanism of isoflurane inhibition. More detailed mechanistic analysis of the mechanisms of isoflurane block will require more refined gating and single-channel studies.

In current models of Na_v gating, the S4 helix, which contains highly conserved positively charged amino acids, plays a critical role as voltage sensor for activation (Chahine et al., 2004). Similar to Na_v, NaChBac undergoes several gating transitions involving gating charge movement (Kuzmenkin et al., 2004), the kinetics of which closely follow those of macroscopic activation. A neutralizing mutation in S4 (R114C) produced similar effects on NaChBac as isoflurane: a positive shift in the voltage dependence of activation and a negative shift in voltage dependence of inactivation (Chahine et al., 2004). The positive shift in the voltage dependence of NaChBac activation and the net decrease in channel conductance by isoflurane suggest a reduction in voltage sensitivity via direct or indirect interaction with the voltage sensor. Because there are no interdomain linkers and the N terminus of NaChBac is too short to achieve N-type tethered ball-chain inactivation, NaChBac is predicted to undergo C-type inactivation (Catterall, 2001; Ren et al., 2001) similar to slow inactivation in Na_v (Balser et al., 1996a; Fukuda et al., 2005) and K_v channels (López-Barneo et al., 1993; Cordero-Morales et al., 2006). This was confirmed by a recent report in which the pore S6 linker was found to be important in NaChBac inactivation (Pavlov et al., 2005). Our results indicate that NaChBac availability was reduced by isoflurane, evident in the negative shift in the inactivation curve as well as marked enhancement of current decay upon opening. The slope factor for inactivation was unaffected, consistent with no effect of isoflurane on the voltage sensor during channel inactivation. Because NaChBac lacks fast inactivation as seen in Na_v, use-dependent block is probably more complicated in Na_v with additional coupling between channel closing and recovery.

There are similarities between the effects of isoflurane and local anesthetics on Na_v. Voltage-dependent block by isoflurane of NaChBac (this study) and of Na_v (Stadnicka et al., 1999; Ouyang et al., 2003; Ouyang and Hemmings, 2005) is consistent with greater anesthetic affinity for the inactivated state, similar to local anesthetic and anticonvulsant block of Na_v (Ragsdale et al., 1994; Nau and Wang, 2004). Multiple state-dependent mechanisms are involved in local anesthetic block. Lidocaine slows gating transitions during activation (Vedantham and Cannon, 1999; Wang et al., 2004), allosterically modifies S4 of domains III and IV (Sheets and Hanck, 2003), and interacts with the extracellular loop of domain IV (E1555; Li et al., 2002) and with the S4 voltage sensor (Hanck et al., 2000; Sheets and Hanck, 2003). These findings and our observation that isoflurane shifts NaChBac activation indicate that isoflurane, similar to local anesthetics, acts via multiple mechanisms to affect channel gating. The outer pore vestibule is a pivotal determinant of local anesthetic binding in Na_v such that the W1531C mutation in domain IV of Na_v1.4 eliminates local anesthetic block (Li et al., 2002; Tsang et al., 2005). Similar target sites for volatile anesthetics could exist in the NaChBac pore region and should be identifiable by site-directed mutagenesis.

Upon prolonged depolarization, mammalian Na_v channels open, and they subsequently enter fast and slow inactivated states (Hille, 2001; Nau and Wang, 2004). The mechanism of slow inactivation differs from that of fast inactivation by involving rearrangement of the selectivity filter (Balser et al., 1996a; Hille, 2001), as seen in C-type inactivation of K_v channels (López-Barneo et al., 1993; Cordero-Morales et al., 2006). According to the modulated receptor model (Hille, 1977; Hondeghem and Katzung, 1977), use-dependent block during repetitive stimulation results from accumulation of fast-inactivated channels and slowed recovery of channel availability from mostly fast inactivation. Local anesthetics also produce use-dependent block by modulating slow inactivation (Balser et al., 1996b; Vedantham and Cannon, 1999; Fukuda et al., 2005) and/or inhibiting activation (Vedantham and Cannon, 1999; Wang et al., 2004). Because the inactivation kinetics of NaChBac resembles C-type inactivation of K_v channels and slow inactivation of Na_v channels, isoflurane might also reduce Na_v availability via effects on activation and inactivation (enhancing inactivation and delaying recovery). The relative contribution of fast and slow inactivation to block of Na_v by isoflurane is an interesting area for further investigation. Structure-function studies to compare volatile anesthetic effects on NaChBac and Na_v channels should provide a useful approach in dissecting the mechanisms of anesthetic modulation of voltage-gated Na⁺ channels.