The effects of general anaesthetics on ligand-gated ion channels

J. P. Dilger

Department of Anesthesiology, State University of New York, Stony Brook, NY 11794-8480, USA

Abstract

Br J Anaesth 2002; 89: 41–51

Keywords: anaesthesia, general; ions, ion channels, ligand-gated; theories of anaesthetic action, molecular

The idea that general anaesthetics produce unconsciousness, analgesia and amnesia by interfering with communication between neurones is conceptually appealing to both scientists and non-scientists. Indeed, there is a long history of considering postsynaptic ligand-gated ion channels (LGICs) as molecular targets for general anaesthetics.59 Advances in experimental techniques, especially in electrophysiology and molecular biology, have fostered a reductionist approach and allowed exploration of the interactions between general anaesthetics and LGICs in increasingly greater molecular detail. Potential binding sites for anaesthetics on some LGICs have now been identified. However, concrete evidence for these sites awaits the determination of the high-resolution structure of LGICs in the absence and presence of anaesthetic.

This review is organized in the following way. First, we give a summary of the structure and function of LGICs, and this is followed by an examination of the ways in which anaesthetics (or any drugs) might affect LGIC function, a critique of the methods used to measure LGIC function, and a compilation of the effects of general anaesthetics (and similar agents such as alcohols) on individual families of LGICs.

Structure of ligand-gated ion channels

Although it was once thought that all LGICs belonged to a single superfamily of channels, it is now clear that there are at least three distinct superfamilies.60 Currently, ion channels are classified according to their topology with respect to the membrane, i.e. the number of membrane-spanning segments and the number of pore loops (Fig. 1).



View larger version (32K):
[in this window]
[in a new window]
 
Fig 1 The structure of ligand-gated ion channels. (A) Structure of the cys-loop superfamily. (B) Structure of the ionotropic glutamate receptor family. (C) Structure of the P2X family of ATP-gated channels.

 
Members of the cys-loop superfamily contain four membrane-spanning segments without any pore loops (Fig. 1A) and are expressed as pentamers.54 The name ‘cys-loop’ refers to the presence of a pair of disulphide-bonded cysteines near the N-terminal of the protein. The muscle-type nicotinic acetylcholine (ACh) receptor (mnAChR) is a heteropentamer with the stoichiometry {alpha}2ß{gamma}{delta} (embryonic or extrajunctional and Torpedo subtype) or {alpha}2ß{epsilon}{delta} (adult or junctional subtype). As viewed from the synaptic cleft, the subunits are arranged clockwise in the sequence {alpha}{gamma}{alpha}ß{delta}. The two ligand binding sites are in the synaptic region at the {alpha}{gamma} ({alpha}{epsilon}) and {alpha}{delta} subunit interfaces.84 These two sites have nearly the same affinity for agonists but antagonists have a higher affinity for the {alpha}{gamma} ({alpha}{epsilon}) site.93 Most, if not all, of the lining of the pore of the channel is provided by second membrane-spanning segment (M2).42 48 The channel is primarily permeable to monovalent cations;1 the permeability of Ca2+ and Mg2+ relative to Na+ is 0.2. Single channels exhibit a linear current– voltage curve with a conductance of 40 pS (embryonic) or 60 pS (adult). The current–voltage curve produced by a large number of mnAChR channels activated by saturating concentrations of ACh has a small degree of rectification because of the weak dependence of channel open-time on voltage.

The basic structural features of mnAChRs (four membrane-spanning segments, ligand-binding sites at subunit interfaces and a pore formed by M2) are thought to be preserved in the other members of the cys-loop superfamily.

Neuronal nicotinic ACh receptors (nnAChR) are formed from either heteropentamers of two {alpha} ({alpha}2–6, {alpha}10) and three ß 2–4) subunits or homopentamers of five {alpha} ({alpha}7–9) subunits. The most common subtype combinations are {alpha}4ß2 (brain), {alpha}3ß4 (sympathetic ganglia) and {alpha}7 (presynaptic terminals).34 nnAChRs are considerably more permeable to divalent cations than are mnAChRs. The permeability of nnAChRs to Ca2+ ranges from 1.5 to 20 times the Na+ permeability.82 This implies that a significant flux of Ca2+ enters the postsynaptic cholinergic neurone during synaptic transmission. nnAChRs exhibit a strongly inwardly rectifying single-channel current–voltage curve with a reversal potential near 0 mV. Rectification results from a voltage-dependent block of the channel47 by intracellular Mg2+. At positive membrane potentials, Mg2+ has high affinity for binding within the pore and blocking the efflux of K+.

Nicotinic ACh receptor (AChR) channels permeable to Cl have been found in molluscan neurones.14 55 However, they have not yet been cloned, so it is not clear how they should be placed in the LGIC classification scheme.

Two types of subunit forming LGICs activated by serotonin (5-hydroxytryptamine or 5-HT) have been identified:20 5-HT3A and 5-HT3B. The 5-HT3A subunit forms homopentameric channels with a very small conductance (<1 pS) that are about 30% more permeable to Ca2+ than Na+. Macroscopic currents display inward rectification.16 When the subunits are co-expressed, the heteropentameric channels that are formed have considerably different properties that more closely resemble native channels in neurones. The single-channel conductance is 16 pS, Ca2+ permeability is about 30% less than Na+ permeability and the current–voltage curve is linear to +40 mV.20

Two members of the cys-loop superfamily, {gamma}-aminobutyric acid type A receptors (GABAAR) and glycine receptors (GlyR), are permeable to anions rather than cations and thus are inhibitory. There are at least 16 subunits in the GABAAR family: {alpha}1–6, ß1–4, {delta}, {epsilon} and {gamma}1–4. An additional set of subunits, {rho}1–3, are now also considered to be GABAC receptors. The most abundant heteropentameric combinations are {alpha}1ß2{gamma}2, {alpha}2ß3{gamma}2 and {alpha}3ßx{gamma}2. Single- channel conductances vary with subtype, but are typically 20–40 pS. The current–voltage relationships at the single-channel and macroscopic levels are nearly linear.73

The GlyR family has five known subunits: {alpha}1–4 and ß.13 The {alpha}2 subunit predominates in neonatal rats, but after birth there is a switch to expression of {alpha}1.69 Alpha subunits can combine to form homopentamers.43 Native GlyR in adult spinal cord is a pentamer of three {alpha}1 and two ß subunits.58 Single-channel conductances are in the range of 10–30 pS with multiple subconductance levels. The current–voltage relationship is linear at the single-channel level but macroscopic currents may be outwardly rectifying as a result of voltage-dependent gating.

The ionotropic glutamate receptor (GluR) superfamily consists of three families. All members are activated in vivo by L-glutamate, but the three families are distinguished by their affinity for the synthetic agonists {alpha}-amino-5-methyl-3-hydroxy-4-isoxazole propionic acid (AMPA), N-methyl-D-aspartame (NMDA) and kainate.68 The membrane topology of a subunit consists of three membrane-spanning segments and one pore-loop sequence (Fig. 1B). Functional receptors are assembled from homo- or heterotetramers.30 AMPA receptors are assembled from either homomers or heteromers of subunits GluR1–4. There are two alternative splice variants of GluR1–4, termed flip and flop; flip is expressed in both embryonic and adult animals whereas flop is expressed almost exclusively in adults. Kainate receptors are assembled from either homomers or heteromers of the subunits GluR5–7 and KA1–2. NMDA receptors are formed from heteromers of NR1 and NR2 (NR2A–D) and sometimes NR3 (NR3A–B). All glutamate receptors are about equally permeable to Na+ and K+. Members of the NMDA family also have a high level of permeability to Ca2+. Single-channel conductances range from 15–75 pS and most channels exhibit subconductance levels. Most GluR channels have an inwardly rectifying current–voltage relationship. For the NMDA family, this results from block by internal Mg2+; for the AMPA and kainate families, it results from block by cytoplasmic polyamine ions.30 AMPA and kainate receptors exhibit desensitization on a timescale of 1–10 ms. Desensitization of NMDA receptors occurs on a timescale of >500 ms.

The ionotropic, purinergic receptor (P2X) ATP-activated superfamily of LGICs exhibit two membrane-spanning regions and no pore-loops (Fig. 1C). Seven subunits have been identified (P2X1–7). They assemble into homomeric and heteromeric trimers80 with at least 11 different combinations.97 All of the receptors are about equally permeable to Na+ and K+ and also have significant Ca2+ permeability.56 Single-channel conductances are of the order of 30 pS and subconductance levels have also been seen. Profound inward rectification has been seen in current–voltage curves from PX2 receptors, which may be attributed to voltage-dependent gating and voltage-dependent single-channel conductance.108

Thus far, none of the LGICs has been crystallized for imaging to atomic resolution. The most detailed structural information about an LGIC comes from electron diffraction images of tubular crystals of the AChR from Torpedo electroplax. The resting state of the channel has been imaged at 4.6 Å resolution77 and differences between the resting and open states have been seen at 9 Å resolution.99 Structural information about the ligand-binding domains of LGICs has been obtained from crystallographic studies. The ligand-binding domain of GluR has been imaged at 1.9 Å resolution to reveal two lobes surrounding a large binding cleft8 in the form of a clamshell. An ACh binding protein was isolated from snail glial cells, purified, crystallized and imaged to 2.7 Å.15 Remarkably, its amino acid sequence resembles the extracellular portion of members of the cys-loop superfamily and it forms a pentamer just like the mammalian channels. Unlike the ligand-binding domain of GluR, there is no large binding cleft in the ACh-binding protein; rather, ligand-binding sites are formed at the interface between each pair of subunits.

Mechanism of action of ligand-gated ion channels

A general scheme for activation and desensitization of LGICs with two agonist binding sites is shown in Figure 2A. Channels can exist in three distinct conformations: resting (R, non-conducting), open (R*, conducting) and desensitized (D, non-conducting). The desensitized state can also be considered as an inactivated state, analogous to voltage-gated ion channels. Sequential binding of two agonist (A) molecules is indicated by the horizontal arrows. Gating refers to transitions between resting and open channels (left-to-right diagonal arrows). Desensitization refers to transitions between resting and desensitized channels (vertical arrows) or open and desensitized channels (right-to-left diagonal arrows). In its completely general form, this nine-state model contains 30 rate constants, of which 23 are independent.



View larger version (22K):
[in this window]
[in a new window]
 
Fig 2 Kinetic schemes used to describe activation and desensitization in ligand-gated ion channels. See text for details. (A) General scheme for ligand-gated channels with two agonist binding sites. (B) The behaviour of mnAChR channels is well described by this part of the general scheme. (C) GABAAR channels have been successfully modelled by this part of the general scheme.

 
The typical scheme used for mnAChR channels is illustrated in Figure 2B. This scheme contains only seven states and nine rate constants (assuming identical ligand binding/dissociation rates in the resting conformation). The simplifications arise from the observations that non-ligand-bound and single-ligand-bound receptors open infrequently,49 that desensitization occurs primarily from the double-ligand-bound open state9 and that recovery from desensitization after removal of agonist occurs primarily through the desensitized pathway.28 When the channel opens, the agonist becomes more tightly bound (indicated by the lack of a pathway from A2R* to AR*) and this increased affinity is preserved in the desensitized state. Many studies ignore (do not measure) the pathway for recovery from desensitization, a simplification that reduces the scheme to five states and six rates. The gating reaction favours the open state by a factor of at least 20, so that, at saturating concentrations of ACh, >95% of the channels open. Gating has weak voltage-dependence; channels close faster at depolarized potentials. Desensitization is fast (time constant ~50 ms) and favours the desensitized state by a factor of >=10, so that 200 ms after addition of ACh most channels are desensitized. After a brief pulse of agonist (as in a synaptic event), desensitization has not begun so the decay of current is governed by channel-closing.

Figure 2C depicts a scheme that has been proposed for the GABAA receptor channel.53 There are seven states and 12 rates (10 after detailed balancing). This scheme differs from that of Figure 2B in several ways. (i) The channel can open to the same conductance level in two distinct ways (other models include a third open state).44 (ii) There are two desensitization pathways (fast and slow); they are accessed through closed states and are voltage-dependent.73 (iii) Recovery from desensitization occurs via the same pathway as activation. (iv) At saturating concentrations of GABA, only about 80% of the channels are open.94 (v) After a brief pulse of agonist, the decay of current is governed by both channel-closing and recovery from desensitization.

How might general anaesthetics affect ligand-gated ion channels?

Before reviewing specific effects of anaesthetics on LGICs, it is useful to consider how anaesthetics might affect channels. The simple answer is that they can either favour open states of the channel (potentiation) or favour closed states of the channel (inhibition). One approach is to simulate this by tweaking the rate constants in the appropriate kinetic scheme (Fig. 2). For example, potentiation might be achieved by increasing agonist affinity, increasing the open gating rate or decreasing desensitization. Inhibition could arise from tweaks in the opposite direction. The process would be repeated for each experimental concentration of anaesthetic and one would end up with a database describing the dependence of rates on anaesthetic concentration. This is a reasonable approach (and perhaps the only approach) if it is thought that anaesthetics act by partitioning into the cell membrane and modifying the physicochemical properties of the membrane. The alterations in channel rate constants would then represent the behaviour of the protein in the new physicochemical lipid environment induced by the anaesthetic. Even if the lipid theory of anaesthetic action were to find experimental support, this approach to understanding the effects of anaesthetics on channels would not be very enlightening. The relationship between rate constants and anaesthetic concentration would be purely empirical. To go beyond empiricism to predictability, it would be necessary to understand the dependence of the membrane property (call it ‘fluidity’, for example) on anaesthetic concentration and then understand how fluidity affects protein conformational changes.

Fortunately, there is much evidence that anaesthetics interact directly with proteins and that this is their primary mode of action. This makes the job of the kinetic modeller simpler: introduce new, anaesthetic-bound states into the model. The underlying assumption is that anaesthetics interact with the receptors in an allosteric fashion.83 Of course, predictability is not guaranteed by this approach because for every additional state there will be additional rate constants to be determined. But you cannot stop a modeller from trying!

Figure 3 shows an example of how an anaesthetic (represented by the letter B) may produce inhibition by binding to the pore of the channel and blocking the flow of current through the pore. The scheme for the mnAChR, omitting desensitization, is used as the starting model. It might be tempting to simply add one additional state to the model, A2R*B, to represent the blocked state. However, this is a restrictive model that implies that the anaesthetic molecule can enter and leave its blocking site only when the gate of the channel is open. The more general model also allows anaesthetic binding to the closed channel conformations and allows that the anaesthetic may bind to these states with different affinities. One testable prediction of this model is that, under conditions where most of the receptors are in the open state (A2R*), occupancy of the open state will be reduced in the presence of anaesthetic by a factor of (1+[B]/KB)–1, where [B] is the anaesthetic concentration and KB is its equilibrium binding constant for the transition A2R*<-->A2R*B. It is sometimes possible to use electrophysiological techniques to determine the binding of anaesthetics to the resting state, but direct binding experiments are often more useful. If the anaesthetic binds preferentially to either the resting or the open state, then agonist binding and/or channel gating must be different for anaesthetic-bound receptors.



View larger version (18K):
[in this window]
[in a new window]
 
Fig 3 A kinetic scheme used to describe inhibition of the mnAChR channel by anaesthetics. See text for details.

 
Figure 4 uses the GABAAR scheme (Fig. 2C) to illustrate how potentiation by anaesthetics (denoted P*) might be modelled. For clarity, states in which the anaesthetic is bound to the open and desensitized forms of the receptor have been omitted. The scheme in Figure 4A shows how the two commonly observed effects of anaesthetics, direct activation and potentiation, may be understood as consequences of a single binding site for the anaesthetic. Direct activation, in the absence of agonist, is represented by the transition R<-->RP*. In principle, the binding affinity of P* to R can be obtained by measuring direct activation as a function of anaesthetic concentration. In practice, it may not be possible to achieve saturating concentrations of the anaesthetic. Potentiation is represented by the binding of P* to the single- and double-ligand-bound resting states. A prediction of this model is that the new conducting states introduced by the anaesthetics should be observed as new components in the open-duration histograms obtained from single-channel recording. In the alternative scheme shown in Figure 4B, the anaesthetic binds preferentially to the double-ligand-bound open state while keeping the channel open. A prediction of this model is that the anaesthetic prolongs one of the open-state components seen in the absence of anaesthetic by a factor of (1+[P]/KP), where [P] is the anaesthetic concentration and KP is its equilibrium binding constant for the transition A2R*<-->A2R*P*.



View larger version (22K):
[in this window]
[in a new window]
 
Fig 4 A kinetic scheme used to describe potentiation of the GABAAR channel by anaesthetics. See text for details.

 
Experimental approaches to the study of ligand-gated ion channels

Some of the techniques used to study LGICs are listed in Table 1. No one technique is ideal. Using intact slice preparations has the advantage of providing a physiological environment for the cells and for observing effects of drugs on synaptic transmission as a whole rather than on one component of the process. In particular, in a slice the time course of neurotransmitter concentration in the synapse is physiological, as are receptor density and the postsynaptic integration of multiple input signals. However, these preparations are not suited to mechanistic studies—there are too many interacting components. Moreover, unless tissues are obtained from transgenic animals, the effects of receptor mutation cannot be studied. Smaller preparations (cells and patches with expressed receptors) provide the experimenter with the control over voltage, concentration and receptor subtype that is necessary for mechanistic studies. The drawbacks of these preparations include varying degrees of artificiality of the conditions and the difficulty of predicting whether drug-induced changes will have a physiological effect in a real synapse or neural circuit.


View this table:
[in this window]
[in a new window]
 
Table 1 A comparison of methods used to study LGICs. Each category is rated good (+), average (0) or poor (–). The categories are as follows: Ex-cell physiol=how well the extracellular environment resembles in vivo conditions; Int-cell physiol=how well the intracellular environment resembles in vivo conditions; synaptic comm=whether synaptic communication is intact; V control=whether voltage-clamp experiments are possible; [ag] control=whether agonist concentration can be experimentally controlled; [ag](t) control=whether the time course of agonist concentration can be controlled; Subunit control=whether signals from different subunit types in the preparation can be distinguished; Mutant control=whether mutant receptors can be studied; Mech info=whether mechanistic information can be obtained from the experiments
 
One experimental factor that is sometimes given inadequate attention in cell and patch studies of LGICs is the speed of the change in agonist concentration used to activate currents. This speed varies with the size of the preparation. Solution exchange times for excised patches can be as fast as 100 µs.63 91 It may be possible to perfuse small cells that are not attached to a substrate within 10 ms.3 However, cells attached to a culture dish may require 200 ms.111 Perfusion of whole oocytes requires several seconds. One consequence of this is illustrated in Figure 5. LGIC currents are assumed to be activated instantaneously and then to decay as a result of desensitization. For the control trace, the desensitization time constant was 0.5 s. A drug is assumed to have two effects on this LGIC: inhibition of the peak current by 50% and a change in the rate of desensitization. For test 1, desensitization is twice as fast as the control and for test 2 it is half as fast as the control. If the system being studied is an excised patch, the true relationships between the test currents and the control can be measured. If the system being studied is an oocyte and the agonist does not equilibrate with the system for, say, 2 s, very different results will be observed. At 2 s, test 1 appears to inhibit the current by 99% and this receptor might be labelled ‘supersensitive’ to the drug. In test 2, this inhibitory drug actually appears to be potentiating the control current. If measurements were made at ~800 ms, this drug might be considered totally ineffective because the test 2 and control currents are the same. Although the patch experiment gives the ‘right’ answer, this may not be the relevant answer for a real synapse because this depends on factors such as the integration time in the postsynaptic cell.



View larger version (17K):
[in this window]
[in a new window]
 
Fig 5 Current simulations illustrating the time-dependence of current modulation by an anaesthetic that has two effects on a channel. The control current desensitizes with a time constant ({tau}) of 0.5 s. For current test 1, the anaesthetic is assumed to decrease the peak amplitude by 50% and increase the rate of desensitization by a factor of 2 ({tau}=0.25 s). For current test 2, the anaesthetic is also assumed to decrease the peak amplitude by 50% but to decrease the rate of desensitization by a factor of 2 ({tau}=1 s). The ratio of test to control is shown in the inset on a logarithmic scale. The degree of inhibition or potentiation observed is critically dependent on the time resolution of the experiment (the first time point that can be measured).

 
As will be shown in the following sections, multiple effects of anaesthetics on LGICs are commonly observed, so the situation illustrated in Figure 5 is not completely theoretical. Clearly, caution must be used in the interpretation of experiments on LGICs when agonist perfusion is slow and desensitization (or some other process) is fast.

Anaesthetics and cys-loop LGICs

Table 2 catalogues the effects of a variety of general anaesthetics on LGICs of the cys-loop superfamily. The purpose of this tabulation is to provide an overview of the effects observed and not to be comprehensive. The many subtypes of heteromeric nnAChRs are not listed separately. A recent review tabulates subtype-specific effects.104 When possible, we note when both inhibitory and potentiating effects have been observed. When the effects are small at minimum alveolar (MA) concentrations, we indicate whether inhibition (i) or potentiation (p) is observed at high concentrations, so that the trend can be noted.


View this table:
[in this window]
[in a new window]
 
Table 2 Effects of anaesthetics on cationic cys-loop LGICs. i=weak inhibition; I=inhibition; II=strong inhibition; p=weak potentiation; P=potentiation; PP=strong potentiation; 0=insensitive
 
Among the muscle-type receptors, inhibition is the most commonly observed effect, but these receptors are relatively insensitive to a number of anaesthetics. Volatile anaesthetics and alcohols have both inhibitory and potentiating effects on mnAChRs.72 Inhibition is manifested in several ways. At the single-channel level, there is a flickering channel behaviour that decreases the open time per burst.26 78 On the macroscopic current level, there is a decrease in the peak current response to saturating concentrations of agonist27 64 and an acceleration of desensitization.27 89 Both channel flickering and the decreased current response can be interpreted in terms of a channel blocking mechanism (Fig. 3) in which the inhibitory binding site is within the channel pore.37 110 Drugs such as isoflurane and butanol have equal affinity for the open and resting states of the channel,27 64 whereas long-chain alcohols, such as octanol, have greater affinity for the open state.37 The molecular site of action for acceleration of desensitization is unknown. The potentiating effects of volatile anaesthetics and alcohols can easily be overlooked. Experiments must be done at low concentrations of ACh or with a partial agonist, such as decamethonium.64 Alternatively, the non-electrophysiological approach of stopped-flow fluorescence spectroscopy90 may be used. The frequency of bursts of single-channel activity increases in the presence of isoflurane26 and many alcohols,64 revealing their potentiating effects. Potentiation is most easily studied with ethanol because it occurs at lower concentrations than inhibition.39 Potentiation by ether64 and isoflurane27 90 arises from the stabilization of ACh binding, whereas potentiation by alcohols39 64 arises from stabilization of the open state. Inhibition of mnAChRs by pentobarbital has also been studied at the mechanistic level.7 24 This barbiturate binds more tightly to the open state of the receptor than to the resting state and acts as a blocker of open channels. Pentobarbital does not accelerate desensitization nor does it have any potentiating effects on the muscle receptor.

The most pronounced effect of anaesthetics on nnAChRs is the inhibition observed with heteromeric receptors in the presence of volatile anaesthetics.36 100 These receptors are sensitive to sub-MA concentrations of volatile anaesthetics and are thus considered to be unimportant for anaesthesia itself. This may be a premature assessment for two reasons. First, most of the reported experiments were done with receptors expressed in oocytes. The poor time resolution of such experiments may provide a distorted picture, especially if the anaesthetics have multiple effects (Fig. 5). Secondly, our knowledge about cholinergic synapses in the CNS is lacking. If there is a large margin of safety at these synapses (as is seen at the neuromuscular junction), it may be necessary to inhibit a large fraction of receptors in order to interfere with synaptic communication.

Potentiating effects by volatile anaesthetics on heteromeric nnAChRs have not been reported. The problem may be that experiments have not been done under conditions that would favour potentiating effects. However, potentiation has been observed with urethane45 and short-chain alcohols.111

Homomeric nnAChRs appear to be relatively insensitive to the anaesthetics that have been tested thus far. 5-HT3Rs exhibit a wide variety of responses to anaesthetics. Although isoflurane, halothane67 and ether109 potentiate these receptors, sevoflurane has mostly inhibitory effects.92 Short-chain alcohols have potentiating and inhibitory effects.50 67 Inhibition is seen with long-chain alcohols,50 barbiturates12 50 and a steroid anaesthetic.12 Propofol inhibits only at high concentrations.12

Effects of anaesthetics on the anionic members of the cys-loop superfamily are listed in Table 3. The neuronal AChR that is permeable to chloride is included in this table, although its superfamily relationship is unknown. In general, family members containing different subunit combinations may be affected to different degrees by the drugs, but for clarity (but not completeness) we have not listed these separately. One important exception is the homomeric GABACR formed from the {rho}1 subunit. This receptor is inhibited rather than potentiated by volatile anaesthetics and alcohols and is not affected by pentobarbital, propofol or alphaxalone.74 This observation prompted the (GABAC {rho}1)–(Gly {alpha}1) chimera receptor experiments that led to the identification of residues that confer sensitivity to potentiation by anaesthetics.76


View this table:
[in this window]
[in a new window]
 
Table 3 Effects of anaesthetics on anionic cys-loop LGICs. For explanation see legend of Table 2
 
GABAARs and GlyRs are potentiated by many, but not all, anaesthetics. The exceptions are nitrous oxide (weak potentiation), xenon (weak potentiation), cyclopropane (no effect at the concentrations studied) and butane (no effect at the concentrations studied). In addition, GlyRs are only weakly potentiated by pentobarbital and are not affected by ketamine or etomidate. Site-directed mutagenesis experiments have localized residues on GABAARs and GlyRs that determine anaesthetic sensitivity.76 The current model envisages a cavity between the membrane-spanning segments of the receptors with Leu232 on M1, Ser270 on M2, and Ala291 on M3 contributing to this cavity.51 This model is discussed in more detail elsewhere in this issue.98

Like the cationic members of the cys-loop superfamily, anaesthetics have multiple actions on GABAA and GlyRs. Direct activation of these channels by some anaesthetics has been observed. Inhibition of the receptors by volatile anaesthetics and barbiturates has been reported to occur at somewhat higher concentrations than those needed to produce potentiation. It is possible that inhibition also occurs with other anaesthetics but the ideal conditions for observing inhibition and potentiation are different, so the proper experiments may not have been performed.

While much of the recent interest on the interactions of anaesthetics with GABAARs has to do with the identification of sites, research specifically on the mechanism of potentiation has been neglected. A notable exception is single-channel studies of the interaction of pentobarbital with expressed GABAARs.5 94 One of these studies addresses the question of whether potentiation results from slowing of agonist dissociation (binding) or from slowing of channel closure (gating). The results are consistent with the idea that pentobarbital stabilizes one of the open states.94

Anaesthetics, ionotropic glutamate ligand-gated ion channels and P2X ATP ligand-gated ion channels

NMDA receptors (NMDA-Rs) are inhibited by many general anaesthetics (Table 4). Of particular interest is the fact that they are strongly inhibited by both nitrous oxide and xenon,40 105 two of the anaesthetics that do not potentiate GABAARs or GlyRs. These recent observations, added to the long-standing observation that ketamine inhibits NMDA-Rs,66 have caused some to speculate that there are (at least) two different routes to anaesthesia, volatile anaesthetics potentiating GABAergic synapses and xenon inhibiting glutamatergic synapses.21 One test of this hypothesis will be to extend measurements on NMDA-Rs to other anaesthetics that do not potentiate GABAARs. Also, it is not clear how the potentiation by volatile anaesthetics of kainate receptors would fit into this scheme.


View this table:
[in this window]
[in a new window]
 
Table 4 Effects of anaesthetics on glutamate-activated and P2X-ATP activated LGICs. For explanation see legend of Table 2
 
The P2X-R is the least comprehensively studied LGIC superfamily when it comes to general anaesthetics (Table 4). Inhibition by volatile anaesthetics, short-chain alcohols and ketamine has been reported.

Summary

The experimental effort that has been expended in investigating the effects of general anaesthetics on LGICs has been enormous over the past decade. Members of all three LGIC superfamilies have been examined using electrophysiological techniques. Anaesthetics that have been examined include volatile anaesthetics, gaseous anaesthetics, alcohols, i.v. anaesthetics and non-immobilizers. Obsolete anaesthetics (ether, cyclopropane, butane) have been used in order to increase the variability of the structure and polarity of experimental compounds. The tools of molecular biology have been used to make chimeric receptors and to make single-site mutations. Interestingly, this work has been taking place in parallel with efforts to understand the structure of these proteins. Anaesthetic research often stimulates structural research as well as vice versa.

There are some common themes in the interactions between anaesthetics and the three superfamilies of LGICs. In many cases, anaesthetics have both inhibitory and potentiating effects on the channels. It is likely that the number of examples of this will increase when experiments are designed to look specifically for one or the other type of effect. So we must conclude that there are multiple binding sites for anaesthetics on LGICs. The degree of inhibition or potentiation is not easily predictable. In retrospect, this is not surprising when we consider that the sensitivity of a channel to anaesthetics can be altered by a single amino-acid mutation. The large structural differences between the cys-loop, glutamate-activated and P2X superfamilies do not lead to large differences in anaesthetic sensitivity. It is the smaller, almost insignificant, changes that do this. This observation that small changes may lead to large effects reinforces the idea that at least some of the interactions between anaesthetics and LGICs are direct drug–protein interactions that are not mediated by the lipids.

This review has not addressed the question of whether the effects of anaesthetics seen on LGICs are relevant to anaesthesia. This question cannot really be answered at present. Although potent effects can be observed on the channels themselves, we have only begun to try to understand whether these effects are important for a synapse, a neuronal circuit or the function of an animal’s nervous system. We have studied the trees; now we must go on to study the forest and the ecosystem.

References

1 Adams DJ, Dwyer TM, Hille B. The permeability of endplate channels to monovalent and divalent metal cations. J Gen Physiol 1980; 75: 493–510[Abstract]

2 Aguayo LG, Pancetti FC. Ethanol modulation of the gamma-aminobutyric acidA- and glycine-activated Cl current in cultured mouse neurons. J Pharmacol Exp Ther 1994; 270: 61–9[Abstract]

3 Akaike N, Inoue M, Krishtal OA. ‘Concentration-clamp’ study of {gamma}-aminobutyric-acid-induced chloride currents in frog sensory neurones. J Physiol (Lond) 1986; 379: 171–85[Abstract]

4 Akaike N, Tokutomi N, Ikemoto Y. Augmentation of GABA-induced current in frog sensory neurons by pentobarbital. J Physiol (Lond) 1990; 258: C452–60

5 Akk G, Steinbach JH. Activation and block of recombinant GABA(A) receptors by pentobarbitone: a single-channel study. Br J Pharmacol 2000; 130: 249–58[Abstract/Free Full Text]

6 Andoh T, Furuya R, Oka K, et al. Differential effects of thiopental on neuronal nicotinic acetylcholine receptors and P2X purinergic receptors in PC12 cells. Anesthesiology 1997; 87: 1199–209[ISI][Medline]

7 Arias HR, McCardy EA, Gallagher MJ, Blanton MP. Interaction of barbiturate analogs with the Torpedo californica nicotinic acetylcholine receptor ion channel. Mol Pharmacol 2001; 60: 497–506.[Abstract/Free Full Text]

8 Armstrong N, Sun Y, Chen GQ, Gouaux E. Structure of a glutamate-receptor ligand-binding core in complex with kainate. Nature 1998; 395: 913–7[ISI][Medline]

9 Auerbach A, Akk G. Desensitization of mouse nicotinic acetylcholine receptor channels. A two-gate mechanism. J Gen Physiol 1998; 112: 181–97[Abstract/Free Full Text]

10 Banks MI, Pearce RA. Dual actions of volatile anesthetics on GABA(A) IPSCs: dissociation of blocking and prolonging effects. Anesthesiology 1999; 90: 120–34[ISI][Medline]

11 Barann M, Bönisch H, Urban BW, Göthert M. Inhibition by propofol of 5-HT3 receptors in excised patches of NIE-115 cells. Ann NY Acad Sci 1998; 861: 278–9[Free Full Text]

12 Barann M, Göthert M, Fink K, Bönisch H. Inhibition by anaesthetics of 14C-guanidinium flux through the voltage-gated sodium channel and the cation channel of the 5-HT3 receptor of N1E-115 neuroblastoma cells. Naunyn Schmiedebergs Arch Pharmacol 1993; 347: 125–32.[ISI][Medline]

13 Betz H, Kuhse J, Schmieden V, Laube B, Kirsch J, Harvey RJ. Structure and functions of inhibitory and excitatory glycine receptors. Ann NY Acad Sci 1999; 868: 667–76[Abstract/Free Full Text]

14 Bregestovksi PD, Bukharaeva EA, Iljin VI. Voltage clamp analysis of acetylcholine receptor desensitization in isolated mollusc neurones. J Physiol (Lond) 1979; 297: 581–95[Abstract]

15 Brejc K, van Dijk WJ, Klaassen RV, et al. Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors. Nature 2001; 411: 269–76[ISI][Medline]

16 Brown AM, Hope AG, Lambert JJ, Peters JA. Ion permeation and conduction in a human recombinant 5-HT3 receptor subunit (h5-HT3A). J Physiol (Lond) 1998; 507: 653–65[Abstract/Free Full Text]

17 Coates KM, Flood P. Ketamine and its preservative, benzethonium chloride, both inhibit human recombinant alpha7 and alpha4beta2 neuronal nicotinic acetylcholine receptors in Xenopus oocytes. Br J Pharmacol 2001; 134: 871–9[Abstract/Free Full Text]

18 Coates KM, Mather LE, Johnson R, Flood P. Thiopental is a competitive inhibitor at the human alpha7 nicotinic acetylcholine receptor. Anesth Analg 2001; 92: 930–3[Abstract/Free Full Text]

19 Cottrell GA, Lambert JJ, Peters JA. Modulation of GABAA receptor activity by alphaxalone. Br J Pharmacol 1987; 90: 491–500[Abstract]

20 Davies PA, Pistis M, Hanna MC, et al. The 5-HT3B subunit is a major determinant of serotonin-receptor function. Nature 1999; 397: 359–63[ISI][Medline]

21 de Sousa SL, Dickinson R, Lieb WR, Franks NP. Contrasting synaptic actions of the inhalational general anesthetics isoflurane and xenon. Anesthesiology 2000; 92: 1055–66.[ISI][Medline]

22 Dildy-Mayfield JE, Eger EI, 2nd, Harris RA. Anesthetics produce subunit-selective actions on glutamate receptors. J Pharmacol Exp Ther 1996; 276: 1058–65[Abstract]

23 Dildy-Mayfield JE, Mihic SJ, Liu Y, Deitrich RA, Harris RA. Actions of long chain alcohols on GABAA and glutamate receptors: relation to in vivo effects. Br J Pharmacol 1996: 118: 378–84[Abstract]

24 Dilger JP, Boguslavsky R, Barann M, Katz T, Vidal AM. Mechanisms of barbiturate inhibition of acetylcholine receptor channels. J Gen Physiol 1997; 109: 401–14[Abstract/Free Full Text]

25 Dilger JP, Brett RS. Actions of volatile anesthetics and alcohols on cholinergic receptor channels. Ann NY Acad Sci 1991; 625: 616–27[ISI][Medline]

26 Dilger JP, Brett RS, Lesko LA. Effects of isoflurane on acetylcholine receptor channels. 1. Single-channel currents. Mol Pharmacol 1992; 41: 127–33[Abstract]

27 Dilger JP, Brett RS, Mody HI. The effects of isoflurane on acetylcholine receptor channels. 2. Currents elicited by rapid perfusion of acetylcholine. Mol Pharmacol 1993; 44: 1056–63[Abstract]

28 Dilger JP, Liu Y. Desensitization of acetylcholine receptors in BC3H-1 cells. Pflügers Arch 1992; 420: 479–85[ISI][Medline]

29 Dilger JP, Vidal AM, Mody HI, Liu Y. Evidence for direct actions of general anesthetics on an ion channel protein—a new look at a unified mechanism of action. Anesthesiology 1994; 81: 431–42[ISI][Medline]

30 Dingledine R, Borges K, Bowie D, Traynelis SF. The glutamate receptor ion channels. Pharmacol Rev 1999; 51: 7–61[Abstract/Free Full Text]

31 Downie DL, Franks NP, Lieb WR. Effects of thiopental and its optical isomers on nicotinic acetylcholine receptors. Anesthesiology 2000; 93: 774–83[ISI][Medline]

32 Downie DL, Hall AC, Lieb WR, Franks NP. Effects of inhalational general anaesthetics on native glycine receptors in rat medullary neurones and recombinant glycine receptors in Xenopus oocytes. Br J Pharmacol 1996; 118: 493–502[Abstract]

33 Downie DL, Vicente-Agullo F, Campos-Caro A, Bushell TJ, Lieb WR, Franks NP. Determinants of the anesthetic sensitivity of neuronal nicotinic acetylcholine receptors. J Biol Chem 2002; 277: 10367–73[Abstract/Free Full Text]

34 Flood P. Effects of volatile anesthetics at nicotinic acetylcholine receptors. In: Moody E, Skolnick P, eds. Molecular Bases of Anesthesia. Boca Raton, FL: CRC Press, 2001; 305–14

35 Flood P, Krasowski MD. Intravenous anesthetics differentially modulate ligand-gated ion channels. Anesthesiology 2000; 92: 1418–25[ISI][Medline]

36 Flood P, Ramirez-Latorre J, Role L. Alpha 4 beta 2 neuronal nicotinic acetylcholine receptors in the central nervous system are inhibited by isoflurane and propofol, but alpha 7-type nicotinic acetylcholine receptors are unaffected. Anesthesiology 1997; 86: 859–65[ISI][Medline]

37 Forman SA, Miller KW, Yellen G. A discrete site for general anesthetics on a postsynaptic receptor. Mol Pharmacol 1995; 48: 574–81[Abstract]

38 Forman SA, Raines DE. Nonanesthetic volatile drugs obey the Meyer–Overton correlation in two molecular protein site models. Anesthesiology 1998; 88: 1535–48[ISI][Medline]

39 Forman SA, Zhou Q. Novel modulation of a nicotinic receptor channel mutant reveals that the open state is stabilized by ethanol. Mol Pharmacol 1999; 55: 102–8[Abstract/Free Full Text]

40 Franks NP, Dickinson R, de Sousa SL, Hall AC, Lieb WR. How does xenon produce anaesthesia? Nature 1998; 396: 324[ISI][Medline]

41 Furuya R, Oka K, Watanabe I, Kamiya Y, Itoh H, Andoh T. The effects of ketamine and propofol on neuronal nicotinic acetylcholine receptors and P2x purinoceptors in PC12 cells. Anesth Analg 1999; 88: 174–80[Abstract/Free Full Text]

42 Giraudat J, Dennis M, Heidmann T, Chang JY, Changeux JP. Structure of the high-affinity binding site for noncompetitive blockers of the acetylcholine receptor: serine-262 of the delta subunit is labeled by [3H]chlorpromazine. Proc Natl Acad Sci USA 1986; 83: 2719–23[Abstract]

43 Griffon N, Buttner C, Nicke A, Kuhse J, Schmalzing G, Betz H. Molecular determinants of glycine receptor subunit assembly. EMBO J 1999; 18: 4711–21[Abstract/Free Full Text]

44 Haas KF, Macdonald RL. GABAA receptor subunit {gamma}2 and {delta} subtypes confer unique kinetic properties on recombinant GABAA receptor currents in mouse fibroblasts. J Physiol (Lond) 1999; 514: 27–45[Abstract/Free Full Text]

45 Hara K, Harris RA. The anesthetic mechanism of urethane: the effects on neurotransmitter-gated ion channels. Anesth Analg 2002; 94: 313–18[Abstract/Free Full Text]

46 Hill-Venning C, Belelli D, Peters JA, Lambert JJ. Subunit-dependent interaction of the general anaesthetic etomidate with the gamma-aminobutyric type A receptor. J Pharmacol 1997; 120: 749–56

47 Ifune CK, Steinbach JH. Rectification of acetylcholine-elicited currents in PC12 pheochromocytoma cells. Proc Natl Acad Sci USA 1990; 87: 4794–8[Abstract]

48 Imoto K, Methfessel C, Sakmann B, et al. Location of a {delta}-subunit region determining ion transport through the acetylcholine receptor channel. Nature 1986; 324: 670–6[ISI][Medline]

49 Jackson MB. Kinetics of unliganded acetylcholine receptor channel gating. Biophys J 1986; 49: 663–72[Abstract]

50 Jenkins A, Franks NP, Lieb WR. Actions of general anaesthetics on 5-HT3 receptors in N1E-115 neuroblastoma cells. Br J Pharmacol 1996; 117: 1507–15[Abstract]

51 Jenkins A, Greenblatt EP, Faulkner HJ, et al. Evidence for a common binding cavity for three general anesthetics within the GABAA receptor. J Neurosci 2001; 21: RC136, 1–4[Abstract/Free Full Text]

52 Jones MV, Brooks PA, Harrison NL. Enhancement of gamma-aminobutyric acid-activated Cl currents in cultured rat hippocampal neurones by three volatile anaesthetics. J Physiol (Lond) 1992; 449: 279–93[Abstract]

53 Jones MV, Westbrook GL. Shaping of IPSCs by endogenous calcineurin activity. J Neurosci 1997; 17: 7626–33[Abstract/Free Full Text]

54 Karlin A. Emerging structure of the nicotinic acetylcholine receptors. Nat Rev Neurosci 2002; 3: 102–14[ISI][Medline]

55 Kehoe J, McIntosh JM. Two distinct nicotinic receptors, one pharmacologically similar to the vertebrate alpha7-containing receptor, mediate Cl currents in Aplysia neurons. J Neurosci 1998; 18: 8198–213[Abstract/Free Full Text]

56 Khakh BS. Molecular physiology of P2X receptors and ATP signaling at synapses. Nat Rev Neurosci 2001; 2: 165–74[ISI][Medline]

57 Lambert JJ, Belelli D, Weir CJ et al. Etomidate: a selective GABAA receptor modulator. MAC2001 conference abstract.

58 Langosch D, Thomas L, Betz H. Conserved quaternary structure of ligand-gated ion channels: the postsynaptic glycine receptor is a pentamer. Proc Natl Acad Sci USA 1988; 85: 7394–8[Abstract]

59 Larrabee MG, Posternack JM. Selective action of anesthetics on synapses and axons in mammalian sympathetic ganglia. J Neurophysiol 1952; 15: 91–114[Free Full Text]

60 Le Novère N, Changeux J-P. LGICdb: the ligand-gated ion channel database. Nucleic Acids Res 2001; 29: 294–5. http://www.pasteur.fr/recherche/banques/LGIC/LGIC.html[Abstract/Free Full Text]

61 Li C, Aguayo L, Peoples RW, Weight FF. Ethanol inhibits a neuronal ATP-gated ion channel. Mol Pharmacol 1993; 44: 871–5[Abstract]

62 Lin LH, Chen LL, Zirrolli JA, Harris RA. General anesthetics potentiate gamma-aminobutyric acid actions on gamma-aminobutyric acidA receptors expressed by Xenopus oocytes: lack of involvement of intracellular calcium. J Pharmacol Exp Ther 1992; 263: 569–78[Abstract]

63 Liu Y, Dilger JP. Opening rate of acetylcholine receptor channels. Biophys J 1991; 60: 424–32[Abstract]

64 Liu Y, Dilger JP, Vidal AM. Effects of alcohols and volatile anesthetics on the activation of nicotinic acetylcholine receptor channels. Mol Pharmacol 1994; 45: 1235–41[Abstract]

65 Macdonald AG, Ramsey RL, Shelton CJ, Usherwood PN. Single channel analysis of ketamine interaction with a quisqualate receptor. Eur J Pharmacol 1992; 210: 223–9[ISI][Medline]

66 MacDonald JF, Bartlett MC, Mody I, et al. Actions of ketamine, phencyclidine and MK-801 on NMDA receptor currents in cultured mouse hippocampal neurones. J Physiol (Lond) 1991; 432: 483–508[Abstract]

67 Machu TK, Harris RA. Alcohols and anesthetics enhance the function of 5-hydroxytryptamine3 receptors expressed in Xenopus laevis oocytes. J Pharmacol Exp Ther 1994; 271: 898–905[Abstract]

68 Madden DR. The structure and function of glutamate receptor ion channels. Nature Rev Neurosci 2002; 3: 91–101[ISI][Medline]

69 Malosio ML, Marqueze-Pouey B, Kuhse J, Betz H. Widespread expression of glycine receptor subunit mRNAs in the adult and developing rat brain. EMBO J 1991; 10: 2401–9[Abstract]

70 Masaki E, Kawamura M, Kato F. Reduction by sevoflurane of adenosine 5'-triphosphate-activated inward current of locus coeruleus neurons in pontine slices of rats. Brain Res 2001; 921: 226–32[ISI][Medline]

71 McKenzie D, Franks NP, Lieb WR. Actions of general anaesthetics on a neuronal nicotinic acetylcholine receptor in isolated identified neurones of Lymnaea stagnalis. Br J Pharmacol 1995; 115: 275–82[Abstract]

72 McLarnon JG, Pennefather P, Quastel DMJ. Mechanisms of nicotinic channel blockade by anesthetics. In: Roth SH, Miller KW, eds. Molecular and Cellular Mechanisms of Anesthetics. New York: Plenum Press, 1986; 155–64

73 Mellor JR, Randall AD. Voltage-dependent deactivation and desensitization of GABA responses in cultured murine cerebellar granule cells. J Physiol (Lond) 1998; 506: 377–90[Abstract/Free Full Text]

74 Mihic SJ, Harris RA. Inhibition of rho1 receptor GABAergic currents by alcohols and volatile anesthetics. J Pharmacol Exp Ther 1996; 277: 411–6[Abstract]

75 Mihic SJ, McQuilkin SJ, Eger EI 2nd, Ionescu P, Harris RA. Potentiation of gamma-aminobutyric acid type A receptor-mediated chloride currents by novel halogenated compounds correlates with their abilities to induce general anesthesia. Mol Pharmacol 1994; 46: 851–7[Abstract]

76 Mihic SJ, Ye Q, Wick MJ, et al. Sites of alcohol and volatile anaesthetic action on GABA(A) and glycine receptors. Nature 1997; 389: 385–9[ISI][Medline]

77 Miyazawa A, Fujiyoshi Y, Stowell M, Unwin N. Nicotinic acetylcholine receptor at 4.6 Å resolution: transverse tunnels in the channel wall. J Mol Biol 1999; 288: 765–86[ISI][Medline]

78 Murrell RD, Braun MS, Haydon DA. Actions of n-alcohols on nicotinic acetylcholine receptor channels in cultured rat myotubes. J Physiol (Lond) 1991; 437: 431–48[Abstract]

79 Narahashi T. Role of neuronal nicotinic acetylcholine receptors in the mechanism of general anesthesia. MAC2001 conference abstract

80 Nicke A, Baumert HG, Rettinger J, et al. P2X1 and P2X3 receptors form stable trimers: a novel structural motif of ligand-gated ion channels. EMBO J 1998; 17: 3016–28[Abstract/Free Full Text]

81 Paradiso K, Sabey K, Evers AS, et al. Steroid inhibition of rat neuronal nicotinic {alpha}4ß2 receptors expressed in HEK293 cells. Mol Pharmacol 2000; 58: 341–51[Abstract/Free Full Text]

82 Patrick J, Sequela P, Vernino S, Amador M, Luetje C, Dani JA. Functional diversity of neuronal nicotinic acetylcholine receptors. Prog Brain Res 1993; 98: 113–20[ISI][Medline]

83 Pearce, RA. The binary elements model: an allosteric approach to receptor gating and drug action. MAC2001 conference abstract

84 Pedersen SE, Cohen JB. d-Tubocurarine binding sites are located at {alpha}{gamma} and {alpha}{delta} subunit interfaces of the nicotinic acetylcholine receptor. Proc Natl Acad Sci USA 1990; 87: 2785–9[Abstract]

85 Peoples RW, White G, Lovinger DM, Weight FF. Ethanol inhibition of N-methyl-aspartate-activated current in mouse hippocampal neurones: whole-cell patch-clamp analysis. Br J Pharmacol 1997; 122: 1035–42[Abstract]

86 Pistis M, Belelli D, Peters JA, Lambert JJ. The interaction of general anaesthetics with recombinant GABAA and glycine receptors expressed in Xenopus laevis oocytes: a comparative study. Br J Pharmacol 1997; 122: 1707–19[Abstract]

87 Raines DE, Claycomb RJ, Forman SA. The actions of nonhalogenated and perhalogenated alkanes on neuronal nicotinic acetylcholine receptors. MAC2001 conference abstract

88 Raines DE, Claycomb RJ, Scheller M, Forman SA. Nonhalogenated alkane anesthetics fail to potentiate agonist actions on two ligand-gated ion channels. Anesthesiology 2001; 95: 470–7[ISI][Medline]

89 Raines DE, Zachariah VT. The alkyl chain dependence of the effect of normal alcohols on agonist-induced nicotinic acetylcholine receptor desensitization kinetics. Anesthesiology 1999; 91: 222–30[ISI][Medline]

90 Raines DE, Zachariah VT. Isoflurane increases the apparent agonist affinity of the nicotinic acetylcholine receptor by reducing the microscopic agonist dissociation constant. Anesthesiology 2000; 92: 775–85[ISI][Medline]

91 Sachs F. Practical limits on the maximal speed of solution exchange for patch clamp experiments. Biophys J 1999; 77: 682–90[Abstract/Free Full Text]

92 Schneider MG, Barann M, Urban BW. Sevoflurane effects on 5-HT3 receptors. MAC2001 conference abstract.

93 Sine SM, Claudio T. {gamma}- and {delta}-subunits regulate the affinity and the cooperativity of ligand binding to the acetylcholine receptor. J Biol Chem 1991; 266: 19369–77[Abstract/Free Full Text]

94 Steinbach JH, Akk G. Modulation of GABA(A) receptor channel gating by pentobarbital. J Physiol (Lond) 2001; 537: 715–33[Abstract/Free Full Text]

95 Suzuki T, Koyama H, Sugimoto M, Uchida I, Mashimo T. The diverse actions of volatile and gaseous anesthetics on human-cloned 5-hydroxytryptamine3 expressed in Xenopus oocytes. Anesthesiology 2002; 96: 699–704[ISI][Medline]

96 Tomioka A, Ueno S, Kohama K, Goto F, Inoue K. Propofol potentiates ATP-activated currents of recombinant P2X(4) receptor channels expressed in human embryonic kidney 293 cells. Neurosci Lett 2000; 284: 167–70[ISI][Medline]

97 Torres GE, Egan TM, Voigt MM. Hetero-oligomeric assembly of P2X receptor subunits. Specificities exist with regard to possible partners. J Biol Chem 1999; 274: 6653–9[Abstract/Free Full Text]

98 Trudell, JR, Bertaccini E. Molecular modelling of specific and non-specific anesthetic interactions. Br J Anaesth 2002; 89: 32–41.[Abstract/Free Full Text]

99 Unwin N. Acetylcholine receptor channel imaged in the open state. Nature 1995; 373: 37–43[ISI][Medline]

100 Violet JM, Downie DL, Nakisa RC, Lieb WR, Franks NP. Differential sensitivities of mammalian neuronal and muscle nicotinic acetylcholine receptors to general anesthetics. Anesthesiology 1997; 86: 866–74[ISI][Medline]

101 Wachtel RE. Ketamine decreases the open time of single-channel currents activated by acetylcholine. Anesthesiology 1988, 68: 563–70

102 Wachtel RE. Relative potencies of volatile anesthetics in altering the kinetics of ion channels in BC3H1 cells. J Pharmacol Exp Ther 1995; 274: 1355–61[Abstract]

103 Wachtel RE, Wegrzynowicz ES. Kinetics of nicotinic acetylcholine ion channels in the presence of intravenous anaesthetics and induction agents. Br J Pharmacol 1992; 106: 623–7[Abstract]

104 Yamakura T, Bertaccini E, Trudell JR, Harris RA. Anesthetics and ion channels: molecular models and sites of action. Annu Rev Pharmacol Toxicol 2001; 41: 23–51[ISI][Medline]

105 Yamakura T, Chavez-Noriega LE, Harris RA. Subunit-dependent inhibition of human neuronal nicotinic acetylcholine receptors and other ligand-gated ion channels by dissociative anesthetics ketamine and dizocilpine. Anesthesiology 2000; 92: 1144–53[ISI][Medline]

106 Yamakura T, Harris RA. Effects of gaseous anesthetics nitrous oxide and xenon on ligand-gated ion channels. Comparison with isoflurane and ethanol. Anesthesiology 2000; 93: 1095–101[ISI][Medline]

107 Yu D, Zhang L, Eisele J-L, Bertrand D, Changeux J-P, Weight FF. Ethanol inhibition of nicotinic acetylcholine type 7 receptors involves the amino-terminal domain of the receptor. Mol Pharmacol 1996; 50: 1010–6[Abstract]

108 Zhou Z, Hume RI. Two mechanisms for inward rectification of current flow through the purinoceptor P2X2 class of ATP-gated channels. J Physiol (Lond) 1998; 507: 353–64[Abstract/Free Full Text]

109 Zhou Q, Lovinger DM. Pharmacologic characteristics of potentiation of 5-HT3 receptors by alcohols and diethyl ether in NCB-20 neuroblastoma cells. J Pharmacol Exp Ther 1996; 278: 732–40[Abstract]

110 Zhou QL, Zhou Q, Forman SA. The n-alcohol site in the nicotinic receptor pore is a hydrophobic patch. Biochemistry 2000; 39: 14920–6[ISI][Medline]

111 Zuo Y, Aistrup GL, Marszalec W, et al. Dual action of n-alcohols on neuronal nicotinic acetylcholine receptors. Mol Pharmacol 2001; 60: 700–11[Abstract/Free Full Text]