Department of Anesthesia, Stanford University School of Medicine, Stanford, CA 94305-5117, USA
Abstract
Br J Anaesth 2002; 89: 91101
Keywords: nerves, motor neurones
Subcortical in vitro studies at the Molecular Mechanisms Conference
There were few presentations at the recent International Conference on Basic and Molecular Mechanisms of Anaesthesia (MAC2001) concerning subcortical anaesthetic actions studied in vitro. In addition to the spinal cord studies from the authors laboratory presented in detail here and the review of the associated literature, there was one other presentation on the spinal cord slice, which will be covered in the appropriate section below (C. Grasshoff and B. Antkowiak, Effect of volatile anaesthetics on action potential firing of spinal neurones in organotypic slice cultures). In a study of the effects of general anaesthetics on thalamic slices (E. Puil, Thalamic effects of general anesthetics: similar and dissimilar actions of isoflurane and barbiturates), Puil found that isoflurane in the medial geniculate body shunts tonic firing of both sodium and calcium channel-mediated bursts by increasing leak conductance. Similar actions were observed for high concentrations of pentobarbital, but low concentrations enhanced excitation and bursting. Pentobarbital actions were not blocked by bicuculline. These results are relevant to the role of the thalamus in mediating arousal to auditory stimuli and its blockade by anaesthetics. In a study on cerebellar slices (M. Debus, K. Unertl and B. Antkowiak, The potency of general anesthetics to suppress spontaneous neuronal activity in cerebellar acute slices compared to neocortical slice cultures), the authors found that a variety of inhalation agents, ethanol and propofol suppressed mean firing rates more potently in neocortical slices than in cerebellar slices by approximately 5-fold, and the potency ratios among the agents were similar for the two preparations. The results are relevant to the actions of anesthetics on locomotion and balance.
The role of the spinal cord in anaesthesia
The spinal cord plays a role in two anaesthetic end-points: analgesia and immobility in response to a noxious stimulus. Recent evidence also suggests that ascending traffic from the spinal cord to the brain may also modulate the effects of anaesthetic agents on arousal.6 9 Analgesia, meaning the loss of painful sensation before loss of consciousness or of immobility, is clearly a property of some agents but may not be of others. In considering analgesic properties with respect to human pain, it is difficult to separate sedation and anxiolysis from analgesia. Animal models of acute pain, measured as tail-flick latency, suggest that inhalation agents are analgesic at sub-immobilizing concentrations62 but at very low concentrations may be hyperalgesic.81 In a model of human experimental pain, isoflurane did not reduce perceived pain intensity.50 In a tail-pressure test, barbiturates and propofol reduce the threshold for withdrawal at levels below those which prevent movement altogether; however, this may represent hyperreflexia rather than enhanced nociception.10 20
Among all the end-points, the most widely employed for inhalation agents is immobility in response to a noxious stimulus. The development of a standard for anaesthetic potency, defined as the minimum alveolar anaesthetic concentration (MAC) required to prevent movement in 50% of subjects in response to a standardized noxious stimulus,19 has facilitated the comparison of potency among inhalation agents and served as the gold standard for anaesthetic concentrations relevant to proposed mechanisms of anaesthesia. Because MAC is determined as the loss of motor response to a noxious stimulus, this anaesthetic action has sometimes been equated with antinociception or analgesia. However, antinociception and immobility are probably distinct from each other. Not only does the reflex threshold increase in human experimental pain while sensation remains unchanged,50 but recent evidence suggests that mouse strains that differ markedly in their responses in pain models43 display only small differences in MAC.63
For many years it was automatically assumed that immobility, like some other anaesthetic end-points, is due to anaesthetic actions in the brain. However, in the early 1990s seminal studies by Rampil52 and Antognini and colleagues1 2 11 clearly showed that the spinal cord is the dominant central nervous system locus for determining the MAC for volatile anaesthetic agents. Thus, for the first time anaesthetic mechanisms can be studied at the site known to be relevant to a given end-point and at the relevant concentrations. The concentrations used in experimental studies should match those required to reach the end-point according to the part of the central nervous system studied, otherwise the results are irrelevant. For instance, sites such as the hippocampus, associated with memory, should be studied at the lower anaesthetic concentrations associated with amnesia rather than at MAC levels. For studies in the spinal cord, MAC equivalents are the appropriate concentrations.
Cellular sites of anaesthetic action within the spinal cord
The spinal components that may contribute to MAC include the central terminations of sensory primary afferent neurones, interneurones of many types, and the cell bodies and initial axon segments of motor neurones. A simplified schematic diagram of the spinal cord circuitry is shown in Figure 1. The peripheral axons of motor neurones distal to the spinal cord probably do not contribute to immobility, nor do actions at the neuromuscular junction.49 Conduction in large-diameter myelinated axons, like that in motor neurones, is not sensitive to anaesthetic agents at concentrations relevant to MAC, as many studies have shown. Moreover, conduction in primary sensory nerve axons is relatively unaffected, although impulse generation at peripheral nociceptors may be modified. There are several studies in vivo demonstrating volatile anaesthetic actions on the receptive field and the sensitivities of dorsal horn interneurones,3 4 7 8 14 32 44 46 64 71 76 77 80 but little has been done in vitro in spinal preparations. Similarly, in vivo studies on the H-reflex document that volatile anaesthetics can depress the simpler monosynaptic circuit from primary afferent nerve terminals to motor neurones. H-reflexes are depressed by volatile agents at subanaesthetic concentrations,26 41 49 but only at high blood concentrations of propofol.36 Etomidate and ketamine actually increase H-reflex amplitude.34 42 The F-wave is an index of motor neurone excitability. The F-wave represents antidromic invasion of central motor neurones that is sufficiently depolarizing to trigger orthodromic impulses after the absolute refractory period has passed. The F-wave may reflect not only intrinsic motor neurone excitability, but also tonic excitatory and inhibitory influences. F-waves are depressed by inhalation agents, including nitrous oxide, at clinically relevant concentrations.5 27 37 49 53 83 Both H- and F-wave depression correlate with loss of movement in response to a noxious stimulus.84
|
This review will focus on anaesthetic actions on motor neurones and on transmitter release from elements presynaptic to them. The hypothesis will be presented that direct actions of anaesthetic agents, in particular volatile agents, on motor neurones contribute importantly to immobility. In this discussion, motor neurones will be treated as a homogeneous population. This is almost certainly an oversimplification as motor neurones vary in size and position according to the muscles they innervateflexors vs extensors, proximal vs distal, and slow postural vs fast movement muscles. Motor units differ in properties such as speed of adaptation to a sustained stimulus. In mammals there is also a separate population of small motor neurones that innervate muscle spindles. The extent to which these populations of motor neurones differ in response to anaesthetic agents is uncertain.
Early studies in the spinal cord
Most early basic neurophysiological studies in vertebrates were performed in spinal cord motor neurones, which became the cell type used most widely for understanding vertebrate neurophysiology.18 Interpretation of experimental results was facilitated by several advantages of the vertebrate spinal cord. The separation of afferent input from motor output in dorsal and ventral roots permitted a defined input by stimulating a dorsal root; extracellular recordings could be made from ventral roots in the knowledge that this reflected the responses of motor neurones.17 After the introduction of microelectrode intracellular recording, the large size of motor neurones made them nearly the only vertebrate neurone that offered advantages comparable to those of the large axons and cell bodies of crustaceans and molluscs. Functionally, a cell impaled by blind probing could be unambiguously identified as a motor neurone by activating it antidromically via the ventral root.
Much of the initial knowledge of anaesthetic actions was developed by gruelling in vivo studies requiring hours of surgical preparation and painstaking maintenance of the experimental animal. In a landmark series of such studies, Somjen demonstrated that ether and thiopental depressed the monosynaptic excitatory postsynaptic potential in all motor neurones, elevated the threshold for impulse initiation in approximately half, and had no consistent effect on resting membrane potential.61 There were no discernible effects on conduction in the primary afferent nerve terminals.60 The results of his studies led to the hypothesis that these two anaesthetics decrease reflex responses predominantly by depressing excitatory synaptic transmission to motor neurones, with an uncertain and variable contribution from an increase in threshold; depression of excitatory synaptic transmission was postulated to be due to postsynaptic changes, although a decrease in excitatory transmitter release could not be excluded.59 With only minor modifications as a result of the proliferation of the receptors and ion channels that are now known to influence neurones, hypotheses such as these still form the nucleus of the debate concerning anaesthetic actions in the spinal cord.
The techniques used in these difficult in vivo studies were superseded in the early 1970s by the in vitro hippocampal slice preparation. The hippocampal slice became the dominant neuronal preparation just in time to participate in and benefit from the development of new knowledge regarding the multiplicity of neurotransmitters and the complexity of determinants of excitability. However, almost simultaneously with the growth of hippocampal slice studies, Otsuka and his colleagues developed an in vitro preparation of the neonatal rat spinal cord,47 which we and others have used extensively in studies of anaesthetic and analgesic actions. Later, Takahashi and Konnerth showed it was possible to study visually identified motor neurones in spinal cord slices, initially by sharp electrode recording65 and eventually with whole-cell patch clamp.38 66 This preparation is now in widespread use.
Factors governing motor neurone excitability
This section outlines receptors and ion channels that determine motor neurone excitability and the anaesthetic actions on them where these are known.
Intrinsic properties
All neurones maintain a resting potential difference between the inside and the outside of the cell membrane. The ultimate cause of this is active ion transport, predominantly of sodium ions from the inside to the outside of the cell, and the passive redistribution of potassium ions to the cell interior. However, the more immediate contributor is the selective potassium permeability of voltage-independent (non-gated) ion channels. Increased activity of these channels, by increasing the conductance to potassium, will move the resting potential towards the potassium equilibrium potential and away from the threshold of impulse initiation. In addition, by lowering membrane resistance these channels will short-circuit excitatory synaptic input and render it less effective. An anaesthetic agent that increased the passive flow of current through such channels would thus reduce excitability. Channels of particular current interest are tandem pore potassium channels. At least some members of this class of channels have a response to volatile anaesthetic agents that would move excitability in this direction.28 79 Halothane and sevoflurane activate a potassium current in hypoglossal motor neurones that has a pH sensitivity similar to that of the TASK-1 tandem pore potassium channel.58
Some types of sodium channels are sensitive to volatile anaesthetic agents at reasonable concentrations.56 A persistent inward sodium current in hypoglossal motor neurones carried through tetrodotoxin-insensitive sodium channels is reduced by halothane.57 Other channels that are active near the resting potential and will modulate excitability are low-voltage-activated (LVA) calcium channels; halothane decreases current through these channels and thus may modulate excitability through an action on them.68 Motor neurones exhibit long-lasting depolarizing plateau potentials that are also due to LVA calcium channels. Some i.v. agents depress plateau potentials and thus decrease excitability.29 In addition, fast transient potassium channels may be activated under certain conditions, causing a delay in reaching the threshold for impulse initiation. However, isoflurane inhibits such channels.23
Motor neurones respond to persistent depolarization above threshold by generating a train of action potentials. The frequency and number of impulses in the train are limited by calcium-dependent potassium channels; alcohols, ketamine and a barbiturate reduce current through channels of this type but volatile anaesthetics exert little effect at anaesthetically relevant concentrations.16
Synaptic inputs: excitatory
Most of the fast ligand-gated excitatory synaptic input to motor neurones is mediated by receptors of the -amino-3-hydroxy-5-methylisoxazolepropionic acid (AMPA) subtype (GluR14); however, at least in the very young rat, spinal cord N-methyl-D-aspartate (NMDA) receptors provide a greater contribution to the depolarization than is observed in hippocampal cells. Although some kainate receptors, particularly GluR5 and KA1, are present on motor neurones,54 kainate receptors provide a vanishingly small contribution to excitatory input (J. Knape and J. J. Kendig, unpublished data). Motor neurones are themselves cholinergic, but there appears to be little contribution from nicotinic receptors to excitatory input; the response to the application of nicotinic agonists is very small (J. Knape and J. J. Kendig, unpublished data). It is not certain whether 5-HT3 receptors are present on motor neurones.
In addition to fast excitatory transmission, there are slower forms of excitatory transmission mediated by metabotropic (G-protein-coupled) receptors of various types. These include serotonin, neurokinin (NK) 1-3, 1- and
2-adrenergic, dopamine, adenosine, cholinergic muscarinic and metabotropic glutamate receptors. The functions of these receptors on motor neurones are not certain, and some may be inhibitory rather than excitatory.54
Synaptic inputs: inhibitory
Unlike other parts of the central nervous system, the major ionotropic inhibitory transmitter in the spinal cord is glycine, although GABA also contributes; there is evidence that the two ligands of inhibitory chloride channels are co-released at the same synaptic terminals.33 Motor neurones are subject to tonic inhibitory input, as evidenced by spontaneous miniature synaptic currents, and also to both feed-forward and feedback inhibition. Feed-forward inhibition permits coordinated movement by enforcing alternate relaxation of flexor and extensor muscles, and prevents the exaggerated reflexes characteristic of some CNS disorders. Feedback inhibition is represented by the classic Renshaw cell, innervated by a collateral branch from the motor neurone axon and in turn forming inhibitory synaptic connections with the motor neurone. The early component of this inhibition is due to glycinergic input and the later component to GABA. In addition to the fast ligand-gated chloride channels, there are other inhibitory inputs to motor neurones mediated by second messenger-linked receptors, including those for GABAB receptors. There is a vast literature on anaesthetic actions on GABAA and glycine receptors, but little on slower modulatory inhibitory channels.
Methods for studying spinal actions of anaesthetics in vitro: extracellular
The sum of the responses of a population of motor neurones can be recorded from an electrode placed over a ventral root in close proximity to the ventral horn, with stimuli applied to a dorsal root. This can be done in vitro with intact spinal cords taken from very young rats. Many of the early studies on anaesthetic actions cited above employed this method in vivo. In addition to the usual advantages of controlled conditions for in vitro recording, the method has the advantages of employing a purely sensory orthodromic volley and the capability of measuring subthreshold motor neurone responses. However, the neuronal circuits still essentially constitute a black box, giving the method the advantage of revealing everything the entire spinal cord can do but the disadvantage, in mechanistic terms, that changes induced by anaesthetics cannot be attributed to one element or another. Extracellular recording, by its nature, does not produce meaningful absolute values for the amplitudes of the responses recorded, so changes must be expressed as percentages of the control amplitude for each experiment. Furthermore, under most experimental conditions only relatively fast and large transients can be observed, again giving limited insight into underlying changes in resting potential or membrane resistance in individual motor neurones. One type of extracellular recording, in which potentials in a ventral root are measured across a sucrose gap of very high resistance, partially resolves these limitations, although again only average changes can be measured, not absolute values of potentials in individual motor neurones. The method thus obscures variability in the population of motor neurones and is influenced predominantly by the largest motor neurones in the pool. However, this method was used in a very influential paper that demonstrated hyperpolarization in motor neurones of the spinal cord of the frog, measured as the algebraic sum of the responses of the total population of motor neurones to a broad range of anaesthetics, including barbiturates, ether and -chloralose.45 The hyperpolarization was not blocked by GABAA antagonists (glycine antagonists were not used) and was hypothesized to be the result of an increase in potassium conductance, because parallel experiments on hippocampal neurones showed hyperpolarization accompanied by a conductance increase.
The monosynaptic reflex and its underlying excitatory postsynaptic potential
Stimulating a lumbar dorsal root and recording from the corresponding ipsilateral ventral root elicits the classic monosynaptic reflex (Fig. 1A), observed as a compound action potential in the ventral nerve root. Thresholds for eliciting this response are so low that little of the underlying excitatory postsynaptic potential (EPSP) can be seen before it is obliterated by the action potential. Offsetting the stimulating and recording electrodes by one or two segments permits a large subthreshold EPSP to be recorded without contamination by the compound action potential. The fast population EPSP in motor neurones is mediated by both AMPA and NMDA receptors; a combination of AMPA-selective and NMDA-selective antagonists almost completely abolishes it. Motor neurones are subject to tonic inhibition via both GABAA and glycine receptors; application of the GABAA antagonist bicuculline or the glycine antagonist strychnine elevates the response.
Anesthetic actions on extracellularly recorded responses are summarized in Table 1. The population EPSP and the monosynaptic reflex are sensitive to volatile anaesthetic agents, including halothane,67 isoflurane,55 enflurane75 and ethanol,74 at concentrations equivalent to MAC or lower. Concentrations that inhibit the monosynaptic reflex correlate well with MAC.70 An experimental anaesthetic, cyclobutane, depresses the monosynaptic reflex, whereas a related non-immobilizing compound does not.35 Alcohols depress both the AMPA and NMDA receptor-mediated components of the EPSP,74 but with some apparent selectivity for the NMDA component. The monosynaptic reflex is not diminished by barbiturates,31 propofol,31 2-adrenergic agonists,55 opiates21 or ketamine.12 30 The population EPSP has not been examined for these agents, but presumably would be little affected. However, depression of the somewhat slower NMDA receptor-mediated component of the EPSP, which one would expect for ketamine, does not necessarily lead to depression of the monosynaptic reflex as the latter is triggered very early in the rising phase of the AMPA component. An early study in cats suggested that propofol, but not ketamine or a barbiturate, depressed monosynaptic reflexes. However, the propofol results were obtained on a background of barbiturate anaesthesia, which might be expected to exert additive effects with propofol.39
|
The slow ventral root potential is a complex polysynaptic response that requires the activation of small-diameter sensory nerves for its full expression (Fig. 1A).40 Because of this property, its evocation by true noxious stimuli and its sensitivity to analgesic agents, it is believed to be related to nociception.48 78 The slow ventral root potential has an early component sensitive to NMDA receptor antagonists and a late component that appears to be mediated by a variety of metabotropic receptors.40
The slow ventral root potential is more sensitive to anaesthetic agents than the fast EPSP. The late metabotropic receptor-mediated component is selectively sensitive to 2-adrenoceptor agonists55 and to opioids,21 whereas volatile anaesthetic agents,55 ketamine,12 midazolam22 and alcohols74 show some selectivity for the early component. It is not certain where the receptors that mediate these anaesthetic actions are located. Certainly there are NMDA receptors on motor neurones, and possibly also metabotropic receptors of various types. However, it is also probable that at least some of the anaesthetic depression of the slow ventral root potential results from anaesthetic actions on interneurones presynaptic to the motor neurones (Fig. 1A). This question cannot be resolved by studies on the intact cord.
Information from genetically engineered mice
Studies on population-evoked ventral root responses in the intact cord are limited in the extent to which mechanistic explanations can be derived. However, some progress can be made. We have investigated responses to enflurane in spinal cords from mice that lack the ß3 subunit of the GABAA receptor,75 which behavioural studies have shown to have an increased enflurane requirement to prevent movement in response to a noxious stimulus.51 However, ventral root responses in spinal cords from the null mutants did not differ from those from cords from wild-type animals in their sensitivity to enflurane. The contributions of GABAA receptors to enflurane actions did change, however; bicuculline significantly attenuated enfluranes depressant actions in cords from wild-type mice but not in mutant cords. The results suggest that EPSP depression results from anaesthetic actions on multiple targets in the spinal cord, and that the mutation caused an unpredicted shift in the proportions of targets in the mix. GABAA receptors became less important, but in order to maintain the same sensitivity other targets must have played a greater role in the mutants than in the wild-type animals. In the same studies, glycine receptors appeared to be up-regulated in the mutants. These receptors also contributed slightly to anaesthetic actions on the EPSP, but to the same extent in wild-type and mutant animals. Thus, some other, unidentified receptors took on a greater role in anaesthetic depression in the mutant animals. The results are theoretically highly interesting, but illustrate the riskiness of assuming that global knockout mutations will provide simple solutions to the puzzle of anaesthetic mechanisms.
Recording from single motor neurones in the spinal cord slice
The disadvantages of extracellular recording are resolved by studies of single motor neurones, most commonly in recent times using patch-clamp techniques to record from cell bodies. The method of preparing slices from the spinal cord, combined with visualization of cells in the cord via infrared illumination on closed circuit TV and fluorescent labelling, allows the identification of large cells in the ventral horn as motor neurones. With a different technique, using antidromic activation via a ventral root, motor neurones can be identified functionally in the intact or semi-intact cord.
When motor neurones are stimulated via a dorsal root (or in slices via an electrode placed in the dorsal root entry zone) (Fig. 1B and C), the changes induced by anaesthetics can be directly related to studies of population-evoked responses in the intact cord, but with greater insight into their mechanistic basis. Furthermore, by evoking responses in motor neurones directly with glutamate, meanwhile blocking presynaptic impulse activity, it is possible to bypass presynaptic elements and unambiguously identify anaesthetic actions on motor neurones themselves (Fig. 1D). As well as advantages, there are some disadvantages to single-cell recording. In the intact cord, finding and identifying motor neurones by blind probing is cumbersome and the yield is low. In slices, visual control makes establishing a recording easy, but with much of the connectivity stripped away important elements may be lost.
Dorsal root-evoked responses are the single-neurone counterpart of the population ventral root responses evoked by dorsal root stimulation in the intact cord. Although it is possible in the intact cord to examine the slow depolarizations that are the counterparts of population slow ventral root potentials, in spinal cord slices any such responses are small because of the diminished input. Most studies have therefore examined relatively fast events, which correspond to the short-latency population EPSP and early polysynaptic responses. In addition to examining excitatory synaptic transmission, single-cell studies using whole-cell patch techniques in current-clamp mode or intracellular sharp electrodes can look at changes in intrinsic excitability by examining resting membrane potential, conductance, threshold and the number of impulses evoked by depolarization to a given level. Studies in voltage-clamp mode can exclude changes in response resulting from resting membrane potential changes, at least in the cell body and proximal dendrites; clamp control in distal dendrites is problematic.
Anaesthetic actions on intrinsic motor neurone excitability
Anaesthetic actions on individual motor neurones are summarized in Table 1. Using intracellular electrodes, Takenoshita and Takahashi67 found that halothane hyperpolarized motor neurone cell membranes in rats by several millivolts while increasing input conductance. The threshold for spike initiation did not change. The authors concluded that the larger current required for impulse initiation was due to the hyperpolarization and the increase in conductance, which they attributed to an increase in potassium permeability. In studies with whole-cell patch electrodes, our studies also showed that halothane consistently hyperpolarized motor neurones, with a reduction in the number of impulses evoked by depolarizing current injections, which could be attributed to the hyperpolarization. However, there was no significant change in input resistance (G. Cheng and J. J. Kendig, unpublished data). Ethanol also decreased the number of impulses evoked by a given level of depolarizing input current, but without changes in resting potential, whereas input resistance increased, as did threshold.73 Enflurane also hyperpolarized mouse spinal cord neurones, but again with no significant change in input resistance.13 In a paper presented at the MAC2001 conference (C. Grasshoff and B. Antkowiak, Effect of volatile anesthetics on action potential firing of spinal neurones in organotypic slice cultures), the spontaneous activity of unidentified neurones induced by low magnesium was also reduced by enflurane, halothane and sevoflurane, Because the studies were extracellular, there is no information on resting potential changes or conductance. The early studies by Somjen cited above suggested that, for ether and thiopental, hyperpolarization was inconsistent. Our studies show that hyperpolarization varies among agents but there is no consistent change in membrane conductance. Thus, reduction in the number of impulses in a train and increase in threshold appear to have different causes for different agents, and our laboratory differs from others in finding no decrease in membrane resistance.
Anaesthetic actions on glutamatergic excitation
As in the intact cord, single-cell studies show that volatile anaesthetic agents and ethanol depress the fast EPSP and the corresponding excitatory postsynaptic current (Fig. 2A).13 67 72 73 Enflurane and ethanol depress both AMPA and NMDA receptor-mediated components of EPSPs and currents.13 72
|
It has been a dominant theory that enhancement of activity at GABAA receptors is an important component, perhaps the only component, of the mechanism of action of a wide variety of anaesthetic agents.24 25 69 Although there has been a recent retreat from the universality of this view,15 it remains pervasive in discussions of molecular mechanisms of anaesthesia. For the spinal cord, similar or greater importance is attached to the enhancement of glycine receptor-mediated inhibition. In order to test whether postsynaptic actions on either or both of these receptors account for anaesthetic depression of glutamate-evoked responses in motor neurones, both receptors were blocked by their respective antagonists. In the presence of bicuculline, strychnine or both, ethanol and enflurane still depressed both AMPA and NMDA types of glutamate-evoked currents (Fig. 1B).13 72 Application of any of the antagonists, alone or in combination, did not change the sensitivity of glutamate-evoked currents to the anaesthetic agents. This result shows that volatile anaesthetics can exert depressant actions on spinal motor neurones independent of actions on either of the inhibitory chloride channels, and equally on currents mediated by both major subtypes of glutamate receptors. The result does not, however, demonstrate that these anaesthetics act directly on the glutamate receptors, as the depression of currents may be mediated indirectly via an action elsewhere.
Actions on glycinergic inhibitory transmission
To define pre- and postsynaptic anaesthetic actions at glycinergic synapses, we investigated the effects of volatile anaesthetic agents on spontaneous and evoked glycinergic currents in spinal cord motor neurones.13a The volatile anaesthetic agents enflurane, isoflurane and halothane significantly increased the frequency of glycinergic miniature inhibitory postsynaptic currents (mIPSCs). However, without tetrodotoxin, isoflurane and halothane had no effect and enflurane decreased spontaneous IPSC frequency. All the anaesthetics prolonged the decay time constant () of both spontaneous and glycine-evoked currents without increasing the amplitude. With tetrodotoxin the total charge transfer was increased, but without tetrodotoxin the charge transfer was unchanged (isoflurane and halothane) or decreased (enflurane). Enflurane-induced increases in mIPSC frequency were not significantly affected by Cd2+ (50 µM), thapsigargin (15 µM), or KB-R7943 (5 µM). KB-R7943 and thapsigargin together abolished the enflurane-induced increase in mIPSC frequency. Thus, there are opposing facilitatory and inhibitory actions of volatile anaesthetics on glycine release that depend on calcium homeostatic mechanisms and sodium channels respectively. Under normal conditions (no tetrodotoxin), the absolute amount of glycinergic inhibition does not increase. The contribution of tonic glycinergic inhibition to anaesthesia may depend on its duration rather than its absolute magnitude.
Comparative studies on glutamate miniature excitatory currents mediated by AMPA receptors show only a decrease in frequency with enflurane whether or not sodium channels are blocked (G. Cheng and J. J. Kendig, unpublished data). Thus, anaesthetic actions on spontaneous transmitter release are transmitter-specific, a finding which implies differences in the presynaptic terminals or in the release machinery for glutamate versus glycine, or in both.
Relative roles of anaesthetic actions on motor neurones and presynaptic elements
Although the studies outlined in the preceding section show clearly that anaesthetic agents can act directly on motor neurones to depress responses to transmitter and to reduce the probability of impulse generation, the methods of isolating postsynaptic responses preclude the assignment of relative sensitivity. Direct application of glutamate does not mimic glutamatergic synaptic transmission precisely. The concentration of glutamate applied does approximate the millimolar concentrations released into the synaptic cleft, but exogenous glutamate also activates extrasynaptic receptors and the time course of action is different. In spite of these limitations, the results do show that motor neurone responses to glutamate are depressed at the same concentrations of anaesthetic that depress circuit-mediated synaptic transmission in the intact cord. Thus, the question of relative importance is like the chicken and the egg. If the motor neurone cannot respond, in a sense it does not matter whether or not presynaptic input is also reduced, and vice versa.
There is a difference between synaptically evoked responses and glutamate-evoked currents with respect to the contributions of GABAA and glycine receptors to anaesthetic actions. Blockade of these inhibitory chloride channels does not alter the anaesthetic sensitivity of motor neurones when the probe is exogenously applied glutamate, but does attenuate anaesthetic actions measured as depression of the monosynaptic EPSP.75 The attenuation is not large, particularly in the case of glycinergic receptors, and clearly the response is still depressed in the presence of antagonists to these receptors. It might be argued that the ability of actions on inhibitory channels to modify glutamate-evoked responses is limited when spontaneous activity is blocked. However, there is considerable tonic release, particularly of glycine, even under these conditions, and both bicuculline and strychnine increase the amplitude of glutamate-evoked currents.
Relationship between studies on motor neurones and movement in vitro
In view of the role of motor neurones as the final common path that integrates all upstream input into a moveno move decision, anaesthetic actions that limit the ability of motor neurones to generate a train of impulses in response to an excitatory transmitter are directly relevant to the anaesthetic end-point of immobility in response to a noxious stimulus. Results in vivo are in agreement with a limited role for inhibitory chloride channels in blocking movement. Block of these channels by intrathecal application of bicuculline and strychnine increases anaesthetic requirement by a maximum of approximately 40%.82
As the reports of the studies cited above make clear, there are multiple targets for anaesthetic actions, both on motor neurones themselves and on elements presynaptic to them. Anaesthetics depress the ability of motor neurones to generate action potentials, by acting on a number of different ion channels that may vary from agent to agent. Some of these may generate hyperpolarization, but this is not universal and is not a prerequisite for reducing impulse generation for at least some agents. The preceding section suggests that a number of voltage-gated and voltage-independent channels regulate excitability. It is probable that not all the targets that modulate intrinsic excitability have been identified.
It is clear that actions of anaesthetic agents on GABAA and glycine receptors are important. However, in particular for volatile agents, they are not the sole or even the major contributors to anaesthetic depression of motor neurone excitability. Actions on these receptors may be more important for some i.v. agents, including propofol and barbiturates, and there is evidence that enhancement of GABA-mediated responses in the spinal cord is more prominent for these agents than for ethanol and isoflurane.14 31 55 74
Actions on the glutamate receptors found on motor neurones are probably important. For ethanol and enflurane there is no apparent selective sensitivity for NMDA vs AMPA currents. Selective actions on NMDA vs AMPA currents have been reported for xenon but not isoflurane in another preparation.15 However, there is still the caution that there is as yet no identified site for volatile agents on the glutamate receptor itself, and anaesthetic actions on receptor currents may be indirect. Ketamine, on the other hand, almost certainly exerts its actions, on motor neurones as elsewhere, predominantly as a non-competitive antagonist at NMDA receptors.
There are large areas to which little attention has been paid but will probably be important in the future. In particular, anaesthetic actions on the intracellular metabolic pathways that lead to changes in receptor phosphorylation constitute an area of research that has hardly been touched. In addition, anaesthetic actions on receptors coupled to G-proteins have not been investigated extensively.
Acknowledgements
Work from the authors laboratory was supported by United States National Institutes of Health grants NS13108 and GM47818 to JJK. I am indebted to staff in E. I. Egers laboratory at the University of California at San Francisco for measuring volatile anaesthetic concentrations and to members of Program Project Group GM47818 for many helpful discussions. However, the opinions expressed herein are solely those of the author. The material presented in this review includes much that was covered in a chapter entitled Spinal cord: anaesthetic actions on motor neurones, which will appear in a book, Neural Mechanisms of Anesthesia, edited by J. Antognini, E. Carstens and D. Raines (Humana Press). Permission has been granted by the publishers to present this material with acknowledgement.
References
1 Antognini JF, Schwartz K. Exaggerated anesthetic requirements in the preferentially anesthetized brain. Anesthesiology 1993; 79: 12449[ISI][Medline]
2 Antognini JF. The relationship among brain, spinal cord and anesthetic requirements. Med Hypotheses 1997; 48: 837[ISI][Medline]
3 Antognini JF, Carstens E, Tabo E, Buzin V. Effect of differential delivery of isoflurane to head and torso on lumbar dorsal horn activity. Anesthesiology 1998; 88: 105561[ISI][Medline]
4 Antognini JF, Carstens E. Increasing isoflurane from 0.9 to 1.1 minimum alveolar concentration minimally affects dorsal horn cell responses to noxious stimulation. Anesthesiology 1999; 90: 20814[ISI][Medline]
5 Antognini JF, Carstens E, Buzin V. Isoflurane depresses motoneuron excitability by a direct spinal action: an F-wave study. Anesth Analg 1999; 88: 6815
6 Antognini JF, Wang XW, Carstens E. Isoflurane action in the spinal cord blunts electroencephalographic and thalamic reticular formation responses to noxious stimulation in goats. Anesthesiology 2000; 92: 55966[ISI][Medline]
7 Antognini JF, Wang XW, Piercy M, Carstens E. Propofol directly depresses lumbar dorsal horn neuronal responses to noxious stimulation in goats. Can J Anaesth 2000; 47: 2739
8 Antognini JF, Chen XG, Sudo M, Sudo S, Carstens E. Variable effects of nitrous oxide at multiple levels of the central nervous system in goats. Vet Res Commun 2001; 25: 52338[ISI][Medline]
9 Antognini JF, Saadi J, Wang XW, Carstens E, Piercy M. Propofol action in both spinal cord and brain blunts electroencephalo graphic responses to noxious stimulation in goats. Sleep 2001; 24: 2631[ISI][Medline]
10 Archer DP, Samanani N, Roth SH. Pentobarbital induces nocifensive hyperreflexia, not hyperalgesia in rats. Can J Anaesth 2000; 47: 68792
11 Borges M, Antognini JF. Does the brain influence somatic responses to noxious stimuli during isoflurane anesthesia? Anesthesiology 1994; 81: 15115[ISI][Medline]
12 Brockmeyer DM, Kendig JJ. Selective effects of ketamine on amino acid-mediated pathways in neonatal rat spinal cord. Br J Anaesth 1995; 74: 7984
13 Cheng G, Kendig JJ. Enflurane directly depresses glutamate AMPA and NMDA currents in mouse spinal cord motor neurons independent of actions on GABAA or glycine receptors. Anesthesiology 2000; 93: 107584[ISI][Medline]
13 Cheng G, Kendig JJ. Pre- and post-synaptic anesthetic actions on glycinergic transmission to spinal cord motor neurons. Br J Pharmacol 2002; in press
14 Collins JG, Kendig JJ, Mason P. Anesthetic actions within the spinal cord: contributions to the state of general anesthesia. Trends Neurosci 1995; 18: 54953[ISI][Medline]
15 de Sousa SL, Dickinson R, Lieb WR, Franks NP. Contrasting synaptic actions of the inhalational general anesthetics isoflurane and xenon. Anesthesiology 2000; 92: 105566[ISI][Medline]
16 Dreixler JC, Jenkins A, Cao YJ, Roizen JD, Houamed KM. Patch-clamp analysis of anesthetic interactions with recombinant SK2 subtype neuronal calcium-activated potassium channels. Anesth Analg 2000; 90: 72732
17 Eccles JC. Synaptic potentials of motor neurons. J Neurophysiol 1946; 9: 87120
18 Eccles JC. The Physiology of Nerve Cells. Baltimore, Maryland: Johns Hopkins Press, 1957
19 Eger EI 2nd, Saidman LJ, Brandstater B. Minimum alveolar anesthetic concentration: a standard of anesthetic potency. Anesthesiology 1965; 26: 75663[ISI][Medline]
20 Ewen A, Archer DP, Samanani N, Roth SH. Hyperalgesia during sedation: effects of barbiturates and propofol in the rat. Can J Anaesth 1995; 42: 53240[Abstract]
21 Feng J, Kendig JJ. Selective effects of alfentanil on nociceptive-related neurotransmission in neonatal rat spinal cord. Br J Anaesth 1995; 74: 6916
22 Feng J, Kendig JJ. Synergistic interactions between midazolam and alfentanil in isolated neonatal rat spinal cord. Br J Anaesth 1996; 77: 37580
23 Franks NP, Lieb WR. Stereospecific effects of inhalational general anesthetic optical isomers on nerve ion channels. Science 1991; 254: 42730[ISI][Medline]
24 Franks NP, Lieb WR. Selective actions of volatile general anaesthetics at molecular and cellular levels. Br J Anaesth 1993; 71: 6576[ISI][Medline]
25 Franks NP, Lieb WR. Molecular and cellular mechanisms of general anaesthesia. Nature 1994; 367: 60714[ISI][Medline]
26 Freund FG, Martin WE, Hornbein TF. The H-reflex as a measure of anesthetic potency in man. Anesthesiology 1969; 30: 6427[ISI][Medline]
27 Friedman Y, King BS, Rampil IJ. Nitrous oxide depresses spinal F waves in rats. Anesthesiology 1996; 85: 13541[ISI][Medline]
28 Gray AT, Winegar BD, Leonoudakis DJ, Forsayeth JR, Yost CS. TOK1 is a volatile anesthetic stimulated K+ channel. Anesthesiology 1998; 88: 107684[ISI][Medline]
29 Guertin PA, Hounsgaard J. Non-volatile general anaesthetics reduce spinal activity by suppressing plateau potentials. Neuroscience 1999; 88: 3538[ISI][Medline]
30 Hao JX, Sjolund BH, Wiesenfeld-Hallin Z. Electrophysiological evidence for an antinociceptive effect of ketamine in the rat spinal cord. Acta Anaesthesiol Scand 1998; 42: 43541[ISI][Medline]
31 Jewett BA, Gibbs LM, Tarasiuk A, Kendig JJ. Propofol and barbiturate depression of spinal nociceptive neurotransmission. Anesthesiology 1992; 77: 114854[ISI][Medline]
32 Jinks S, Antognini JF, Carstens E, Buzin V, Simons C. Isoflurane can indirectly depress lumbar dorsal horn activity in the goat via action within the brain. Br J Anaesth 1999; 82: 2449
33 Jonas P, Bischofberger J, Sandkuhler J. Corelease of two fast neurotransmitters at a central synapse. Science 1998; 281: 41924
34 Kano T, Shimoji K. The effects of ketamine and neuroleptanalgesia on the evoked electrospinogram and electromyogram in man. Anesthesiology 1974; 40: 2416[ISI][Medline]
35 Kendig JJ, Kodde A, Gibbs LM, Ionescu P, Eger EI. Correlates of anesthetic properties in isolated spinal cord: cyclobutanes. Eur J Pharmacol 1994; 264: 42736[ISI][Medline]
36 Kerz T, Hennes HJ, Feve A, Decq P, Filipetti P, Duvaldestin P. Effects of propofol on H-reflex in humans. Anesthesiology 2001; 94: 327[ISI][Medline]
37 King BS, Rampil IJ. Anesthetic depression of spinal motor neurons may contribute to lack of movement in response to noxious stimuli. Anesthesiology 1994; 81: 148492[ISI][Medline]
38 Konnerth A, Keller BU, Lev-Tov A. Patch clamp analysis of excitatory synapses in mammalian spinal cord slices. Pflugers Arch 1990; 417: 28590[ISI][Medline]
39 Lodge D, Anis NA. Effects of ketamine and three other anaesthetics on spinal reflexes and inhibitions in the cat. Br J Anaesth 1984; 56: 114351[Abstract]
40 Lozier AP, Kendig JJ. Long-term potentiation in an isolated peripheral nervespinal cord preparation. J Neurophysiol 1995; 74: 10019
41 Mavroudakis N, Vandesteene A, Brunko E, Defevrimont M, Zegers de Beyl D. Spinal and brain-stem SEPs and H reflex during enflurane anesthesia. Electroencephalogr Clin Neurophysiol 1994; 92: 825[ISI][Medline]
42 Meinck HM, Mohlenhof O, Kettler D. Neurophysiological effects of etomidate, a new short-acting hypnotic. Electroencephalogr Clin Neurophysiol 1980; 50: 51522[ISI][Medline]
43 Mogil JS, Wilson SG, Bon K, et al. Heritability of nociception I: responses of 11 inbred mouse strains on 12 measures of nociception. Pain 1999; 80: 6782[ISI][Medline]
44 Namiki A, Collins JG, Kitahata LM, Kikuchi H, Homma E, Thalhammer JG. Effects of halothane on spinal neuronal responses to graded noxious heat stimulation in the cat. Anesthesiology 1980; 53: 47580[ISI][Medline]
45 Nicoll RA, Madison DV. General anesthetics hyperpolarize neurons in the vertebrate central nervous system. Science 1982; 217: 10557[ISI][Medline]
46 Ota K, Yanagidani T, Kishikawa K, Yamamori Y, Collins JG. Cutaneous responsiveness of lumbar spinal dorsal horn neurons is reduced by general anesthesia, an effect dependent in part on GABAA mechanisms. J Neurophysiol 1998; 80: 138390
47 Otsuka M, Konishi S. Electrophysiology of mammalian spinal cord in vitro. Nature 1974; 252: 7334[ISI][Medline]
48 Otsuka M, Yanagisawa M. Effect of a tachykinin antagonist on a nociceptive reflex in the isolated spinal cord-tail preparation of the newborn rat. J Physiol (Lond) 1988; 395: 25570[Abstract]
49 Pereon Y, Bernard JM, Nguyen The Tich S, Genet R, Petitfaux F, Guiheneuc P. The effects of desflurane on the nervous system: from spinal cord to muscles. Anesth Analg 1999; 89: 4905
50 Petersen-Felix S, Arendt-Nielsen L, Bak P, et al. Analgesic effect in humans of subanaesthetic isoflurane concentrations evaluated by experimentally induced pain. Br J Anaesth 1995; 75: 5560
51 Quinlan JJ, Homanics GE, Firestone LL. Anesthesia sensitivity in mice that lack the beta3 subunit of the gamma-aminobutyric acid type A receptor. Anesthesiology 1998; 88: 77580[ISI][Medline]
52 Rampil IJ. Anesthetic potency is not altered after hypothermic spinal cord transection in rats. Anesthesiology 1994; 80: 60610[ISI][Medline]
53 Rampil IJ, King BS. Volatile anesthetics depress spinal motor neurons. Anesthesiology 1996; 85: 12934[ISI][Medline]
54 Rekling JC, Funk GD, Bayliss DA, Dong XW, Feldman JL. Synaptic control of motoneuronal excitability. Physiol Rev 2000; 80: 767852
55 Savola MK, Woodley SJ, Maze M, Kendig JJ. Isoflurane and an alpha 2-adrenoceptor agonist suppress nociceptive neurotransmission in neonatal rat spinal cord. Anesthesiology 1991; 75: 48998[ISI][Medline]
56 Scholz A, Appel N, Vogel W. Two types of TTX-resistant and one TTX-sensitive Na+ channel in rat dorsal root ganglion neurons and their blockade by halothane. Eur J Neurosci 1998; 10: 254756[ISI][Medline]
57 Sirois JE, Pancrazio JJ, Lynch C III, Bayliss DA. Multiple ionic mechanisms mediate inhibition of rat motoneurones by inhalation anaesthetics. J Physiol (Lond) 1998; 512: 85162
58 Sirois JE, Lei Q, Talley EM, Lynch C 3rd, Bayliss DA. The TASK-1 two-pore domain K+ channel is a molecular substrate for neuronal effects of inhalation anesthetics. J Neurosci 2000; 20: 634754
59 Somjen G. Effects of anesthetics on the spinal cord of mammals. Anesthesiology 1967; 28: 13543[ISI][Medline]
60 Somjen GG. Effects of ether and thiopental on spinal presynaptic terminals. J Pharmacol Exp Ther 1963; 140: 395402
61 Somjen GG, Gill M. The mechanism of the blockade of synaptic transmission in the mammalian spinal cord by diethyl ether and by thiopental. J Pharmacol Exp Ther 1963; 140: 1930[ISI]
62 Sonner J, Li J, Eger EI 2nd. Desflurane and nitrous oxide, but not nonimmobilizers, affect nociceptive responses. Anesth Analg 1998; 86: 62934[Abstract]
63 Sonner JM, Gong D, Li J, Eger EI 2nd, Laster MJ. Mouse strain modestly influences minimum alveolar anesthetic concentration and convulsivity of inhaled compounds. Anesth Analg 1999; 89: 10304
64 Sudo M, Sudo S, Chen XG, Piercy M, Carstens E, Antognini JF. Thiopental directly depresses lumbar dorsal horn neuronal responses to noxious mechanical stimulation in goats. Acta Anaesthesiol Scand 2001; 45: 8239[ISI][Medline]
65 Takahashi T. Intracellular recording from visually identified motoneurons in rat spinal cord slices. Proc R Soc Lond B Biol Sci 1978; 202: 41721[ISI][Medline]
66 Takahashi T. Membrane currents in visually identified motoneurones of neonatal rat spinal cord. J Physiol (Lond) 1990; 423: 2746[Abstract]
67 Takenoshita M, Takahashi T. Mechanisms of halothane action on synaptic transmission in motoneurons of the newborn rat spinal cord in vitro. Brain Res 1987; 402: 30310[ISI][Medline]
68 Takenoshita M, Steinbach JH. Halothane blocks low-voltage-activated calcium current in rat sensory neurons. J Neurosci 1991; 11: 140412[Abstract]
69 Tanelian DL, Kosek P, Mody I, MacIver MB. The role of the GABAA receptor/chloride channel complex in anesthesia. Anesthesiology 1993; 78: 75776[ISI][Medline]
70 Tsutahara S, Furumido H, Ohta Y, Harasawa K, Yamamura T, Kemmotsu O. [Effects of halothane, isoflurane, enflurane, and sevoflurane on the monosynaptic reflex response in the isolated spinal cord of newborn rats.] Masui 1996; 45: 82936[Medline]
71 Uchida H, Kishikawa K, Collins JG. Effect of propofol on spinal dorsal horn neurons. Comparison with lack of ketamine effects. Anesthesiology 1995; 83: 131222[ISI][Medline]
72 Wang MY, Rampil IJ, Kendig JJ. Ethanol directly depresses AMPA and NMDA glutamate currents in spinal cord motor neurons independent of actions on GABAA or glycine receptors. J Pharmacol Exp Ther 1999; 290: 3627
73 Wang MY, Kendig JJ. Patch clamp studies of motor neurons in spinal cord slices: a tool for high-resolution analysis of drug actions. Acta Pharmacol Sin 2000; 21: 481576[ISI][Medline]
74 Wong SM, Fong E, Tauck DL, Kendig JJ. Ethanol as a general anesthetic: actions in spinal cord. Eur J Pharmacol 1997; 329: 1217[ISI][Medline]
75 Wong SME, Cheng G, Homanics G, Kendig JJ. Enflurane actions on spinal cords from mice that lack the beta3 subunit of the GABAA receptor. Anesthesiology 2001; 95: 15464[ISI][Medline]
76 Yamamori Y, Kishikawa K, Collins JG. Halothane effects on low-threshold receptive field size of rat spinal dorsal horn neurons appear to be independent of supraspinal modulatory systems. Brain Res 1995; 702: 1628[ISI][Medline]
77 Yanagidani T, Ota K, Collins JG. Complex effects of general anesthesia on sensory processing in the spinal dorsal horn. Brain Res 1998; 812: 3014[ISI][Medline]
78 Yanagisawa M, Murakoshi T, Tamai S, Otsuka M. Tail-pinch method in vitro and the effects of some antinociceptive compounds. Eur J Pharmacol 1984; 106: 2319[ISI][Medline]
79 Yost CS, Gray AT, Winegar BD, Leonoudakis D. Baseline K+ channels as targets of general anesthetics: studies of the action of volatile anesthetics on TOK1. Toxicol Lett 1998; 100101: 293300
80 Zhang C, Davies MF, Guo TZ, Maze M. The analgesic action of nitrous oxide is dependent on the release of norepinephrine in the dorsal horn of the spinal cord. Anesthesiology 1999; 91: 14017[ISI][Medline]
81 Zhang Y, Eger EI, Dutton RC, Sonner JM. Inhaled anesthetics have hyperalgesic effects at 0.1 minimum alveolar anesthetic concentration. Anesth Analg 2000; 91: 4626
82 Zhang Y, Wu S, Eger EI, Sonner JM. Neither GABA(A) nor strychnine-sensitive glycine receptors are the sole mediators of MAC for isoflurane. Anesth Analg 2001; 92: 1237
83 Zhou HH, Mehta M, Leis AA. Spinal cord motoneuron excitability during isoflurane and nitrous oxide anesthesia. Anesthesiology 1997; 86: 3027[ISI][Medline]
84 Zhou HH, Jin TT, Qin B, Turndorf H. Suppression of spinal cord motoneuron excitability correlates with surgical immobility during isoflurane anesthesia. Anesthesiology 1998; 88: 95561[ISI][Medline]