Membrane interactions with general and local anaesthetics: a review of molecular hypotheses of anaesthesia

High-pressure crystallography shows noble gas intervention into protein-lipid interaction and suggests a model for anaesthetic action

In this work we examine how small hydrophobic molecules such as inert gases interact with membrane proteins (MPs) at a molecular level. High pressure atmospheres of argon and krypton were used to produce noble gas derivatives of crystals of three well studied MPs (two different proton pumps and a sodium light-driven ion pump). The structures obtained using X-ray crystallography showed that the vast majority of argon and krypton binding sites were located on the outer hydrophobic surface of the MPs – a surface usually accommodating hydrophobic chains of annular lipids (which are known structural and functional determinants for MPs). In conformity with these results, supplementary in silico molecular dynamics (MD) analysis predicted even greater numbers of argon and krypton binding positions on MP surface within the bilayer. These results indicate a potential importance of such interactions, particularly as related to the phenomenon of noble gas-induced anaesthesia.

Noble gases are known to interact with proteins and can be good anaesthetics in hyperbaric conditions. This study identifies argon and krypton binding sites on membrane proteins and proposes as a hypothesis that noble gases, by altering protein/lipid contacts, may affect protein function.

Influence of trifluoroethanol on membrane interfacial anchoring interactions of transmembrane alpha-helical peptides

Interfacial anchoring interactions between aromatic amino acid residues and the lipid-water interface are believed to be important determinants for membrane protein structure and function. Thus, it is possible that molecules that partition into the lipid-water interface can influence membrane protein activity simply by interfering with these anchoring interactions. Here we tested this hypothesis by investigating the effects of 2,2,2-trifluoroethanol (TFE) on the interaction of a Trp-flanked synthetic transmembrane peptide (acetyl-GW(2)(LA)(8)LW(2)A-NH(2)) with model membranes of dimyristoylphosphatidylcholine. Two striking observations were made. First, using (2)H nuclear magnetic resonance on acyl chain deuterated lipids, we found that addition of 4 or 8 vol % of TFE completely abolishes the ability of the peptide to order and stretch the lipid acyl chains in these relatively thin bilayers. Second, we observed that addition of 8 vol % TFE reduces the tilt angle of the peptide from 5.3 degrees to 2.5 degrees, as measured by (2)H NMR on Ala-d(4) labeled peptides. The "straightening" of the peptide was accompanied by an increased exposure of Trp to the aqueous phase, as shown by Trp-fluorescence quenching experiments using acrylamide. The observation of a reduced tilt angle was surprising because we also found that TFE partioning results in a significant thinning of the membrane, which would increase the extent of hydrophobic mismatch. In contrast to the Trp-flanked peptide, no effect of TFE was observed on the interaction of a Lys-flanked analog (acetyl-GK(2)(LA)(8)LK(2)A-NH(2)) with the lipid bilayer. These results emphasize the importance of interfacial anchoring interactions for membrane organization and provide new insights into how molecules such as TFE that can act as anesthetics may affect the behavior of membrane proteins that are enriched in aromatic amino acids at the lipid-water interface.

Involvement of the extracellular signal regulated kinase pathway in hydrocarbon-induced reactive oxygen species formation in human neutrophil granulocytes

In the present study we have examined the effects of hydrocarbons on the formation of reactive oxygen species (ROS) in human neutrophil granulocytes in vitro. We found that hydrocarbons induce ROS formation in a concentration-dependent manner and that the ROS-inducing potency increases with increasing number of carbon atoms in the structure. In general, aromatic hydrocarbons were less potent inducers of ROS than aliphatic and cyclic hydrocarbons. The most potent compound in each group, t-butylcyclohexane, n-decane, and n-butylbenzene, were chosen for mechanistic studies. ROS formation was inhibited by the MEK1/2 inhibitor U0126, the tyrosine kinase inhibitor erbstatin-A, and the phosphatidylinositol-3 kinase inhibitor wortmannin. The involvement of the ERK1/2 pathway was confirmed by Western blot analysis of phosphorylated ERK1/2. The study revealed only small differences in the mechanisms involved for the three compounds. The responses were not affected by Pertussis toxin, indicating that Gi-protein coupled receptors are not involved in neutrophil activation after hydrocarbon exposure. Based on these findings we propose a mechanism involving tyrosine kinases, PI3 kinase, and the ERK1/2 pathway, leading to activation of the NADPH oxidase and production of ROS in neutrophils stimulated by organic solvents.

The risk of nosocomial pneumonia is not increased during partial liquid ventilation

OBJECTIVE

To determine whether partial liquid ventilation (PLV) affects the risk of nosocomial pneumonia.

STUDY DESIGN

To assess in vitro bacterial adhesion and viability after liquid perfluorocarbon exposure and to assess bacterial recovery after partial liquid ventilation in vivo in rabbits.

SETTING

University animal research facility.

SUBJECTS

Thirty-six New Zealand White rabbits.

INTERVENTIONS

To assess adhesions, radiolabeled Escherichia coli were exposed to perfluorocarbon, incubated against artificial biosurfaces, and compared with nonexposed controls. Bacterial viability in vitro was assessed by exposing broth suspensions of Pasteurella multocida to perflubron for various times. Controls were run in parallel without exposure. Quantitative cultures were performed to determine viability. We undertook short-term and recovery in vivo investigations. The lungs of treated animals were filled with perflubron (approximately 18 mL/kg), and the control rabbits were ventilated without perflubron in an identical fashion. Cryopreserved aliquots of P. multocida were administered via an endotracheal tube. The short-term study animals were ventilated for 6 hrs before being killed. The recovery animals were ventilated for 2-4 hrs, extubated, and killed 20 hrs later. The lungs were removed, aseptically minced, and homogenized. Serial dilutions of the homogenate were quantitatively cultured by manual counting of colonies on agar plates. The recovered organisms were typed for species by the clinical microbiology laboratory.

MEASUREMENTS AND MAIN RESULTS

The adhesion of bacteria to immobilized bronchoalveolar lavage and human saliva, respectively, was reduced by 65%+/-7% and 66%+/-1% (p < .05; n = 5) after exposure to perflubron and by 63%+/-9% and 68%+/-6% after exposure to FC-77 (p < .05; n = 5); however, adhesion was not affected by exposure to Rimar. There was no difference in bacterial viability between the control and perflubron-exposed bacteria (n = 5). The in vivo study demonstrated a ten-fold or greater reduction in the number of recovered bacteria in the partial liquid ventilated group compared with the control group.

CONCLUSIONS

This study suggests that different perfluorocarbons affect adhesions differently. Perflubron and FC-77 appear to decrease bacterial adhesion, whereas Rimar does not. Rerflubron does not have a direct bactericidal effect. Furthermore, PLV with perflubron decreased the number of viable bacteria per gram of tissue after an intentional inoculation of the airway, suggesting that the risk of nosocomial pneumonia is unlikely to be increased during PLV and may, in fact, be reduced in patients supported with PLV.

Liquid ventilation attenuates pulmonary oxidative damage

PURPOSE

Liquid perfluorochemicals reduce the production of reaction oxygen species by alveolar macrophages. We sought to determine whether the use of liquid perfluorochemicals in vivo during liquid ventilation would attenuate oxidative damage to the lung.

MATERIALS AND METHODS

Healthy infant piglets (n = 16) were instrumented for mechanical ventilation and received intravenous oleic acid to create an acute lung injury. The animals were assigned to a nontreatment group receiving conventional mechanical ventilation or a treatment group receiving partial liquid ventilation with a liquid perfluorochemical. Following sacrifice, the bronchoalveolar lavage and lung parenchyma were analyzed for evidence of oxidative damage to lipids and proteins by determination of TBARS and carbonylated protein residues, respectively.

RESULTS

Mortality in the control group was 50% at the completion of the study compared with no deaths in the partial liquid ventilation group (P = .025). The alveolar-arterial oxygen difference was more favorable following injury in the partial liquid ventilation group. The liquid ventilation group demonstrated a 32% reduction in TBARS (P = .043) and a 14% reduction in carbonylated protein residues (P = .061).

CONCLUSION

These data suggest that partial liquid ventilation supports gas exchange and reduces mortality in association with a reduction in the production of reactive oxygen species and the concomitant attenuation of tissue damage during the early phase of acute lung injury.

The internal critical level concept of nonspecific toxicity

The internal lethal concentration (ILC) can be an effective approach in describing the toxicity of a chemical to aquatic organisms that can complement the use of the external toxic concentration characteristic of the LC50. The ILC is an estimate of the toxicant concentration close to the target site and can be estimated from bioconcentration relationships and acute toxicity data. The observed ILC values were found to be consistent for organic compounds exerting the same mode of toxic action. The nonspecific toxicants have the lowest toxicity and the highest ILC values, whereas the chemicals exhibiting specific modes of action have lower concentrations and higher toxicity. There are some reports that the ILC value decreases with increasing exposure periods for various organic chemicals with aquatic organisms. The nonspecific toxicants possibly exhibit their toxic action at the target site by at least two different mechanisms depending on the toxicant concentrations. First, the toxicants bind directly to membrane proteins at relatively low concentrations, resulting in reversible toxic effect. Second, the toxicants inhibit the membrane proteins, and alterations in the lipid bilayers occur at toxicant concentrations sufficient to produce mortality of the organisms. The nonspecific toxicity expressed as acute and chronic toxicity measures are found to correlate well with log Kow. However, the relationship between the ILC and log Kow is less satisfactory because the values of ILC are relatively consistent compared to those of LC50.

Partial liquid ventilation enhances surfactant phospholipid production

OBJECTIVE

To study the effect of partial liquid ventilation on phospholipid metabolism.

DESIGN

Prospective, controlled laboratory study.

SETTING

University-affiliated animal research facility.

SUBJECTS

Mature New Zealand white rabbits (n = 17).

INTERVENTIONS

The rabbits were sedated, anesthetized, and instrumented with tracheostomy and the insertion of an arterial catheter. The rabbits were sequentially assigned to receive conventional mechanical ventilation or partial liquid ventilation with Perflubron (18 mL/kg by bolus fill). Ventilator strategies were identical in both groups and consisted of an FiO2 of 0.5, positive end-expiratory pressure of 4 cm H2O, effective tidal volume of 8 to 13 mL/kg, and rate to maintain Pco2 of 30 to 40 torr (4.0 to 5.3 kPa). Phosphatidylcholine was labeled in vivo by injection of 3H-methylcholine (25 microCi/kg iv). Ventilation was continued for 5.5 hrs.

MEASUREMENTS AND MAIN RESULTS

When animals were killed, phosphatidylcholine was extracted from the total lung lavage and from the pulmonary parenchyma. After the separation of phospholipids by thin-layer chromatography, the 3H activity was determined by liquid scintillation counting. Inorganic phosphorus was also determined to assess the enrichment of the phosphatidylcholine. The 3H-phosphatidylcholine activity in the partial liquid ventilation treated- vs. control rabbits demonstrated a 53% increase (p = .051) in the lavage and a 48% increase (p = .013) in the parenchyma for a net 50% (p = .012) total pulmonary increase. The phospholipid content of the partial liquid ventilation treated- vs. the control rabbits demonstrated a 78% increase (p = .046).

CONCLUSIONS

We conclude that partial liquid ventilation with Perflubron appears to have no negative impact on phospholipid metabolism but rather enhances surfactant phospholipid synthesis and secretion.

Effects of halothane on the nicotinic acetylcholine receptor from Torpedo californica

To determine whether the binding of anesthetics to key membrane receptors is a plausible mode of action, we modeled the effect of the general anesthetic halothane in the nicotinic acetylcholine receptor membrane system isolated from Torpedo californica. Our results demonstrated that halothane inhibits the binding of [3H]phencyclidine ([3H]PCP) to the acetylcholine receptor. The inhibition was reversible, concentration dependent, and had an equilibrium dissociation constant (Kd) of 2.2% atm halothane at 25 degrees. Double-reciprocal plots of the halothane effects at various phencyclidine (PCP) concentrations imply that, under equilibrium conditions, halothane inhibits [3H]PCP binding competitively. In contrast, results from kinetic studies showed that the rate of PCP dissociation is highly sensitive to halothane with EC50 = 0.8% atm halothane in nitrogen. Several possible interpretations are discussed; however, the basic observation was that the kinetics of [3H]PCP binding to the nicotinic acetylcholine receptor was affected by halothane at low concentrations in this model system.

Can the stereoselective effects of the anesthetic isoflurane be accounted for by lipid solubility?

Isoflurane is an inhalational general anesthetic widely used in surgical operations as a racemic mixture of its two optical isomers. The recent availability of pure enantiomers of isoflurane has encouraged their use in experimental studies, and stereoselective effects have now been observed on anesthetic-sensitive neuronal ion channels. Although it has been assumed that such chiral effects demonstrate direct interactions with proteins, it is possible that they could be due to stereoselective interactions with chiral membrane lipids. We have determined the partition coefficients of the two optical isomers of isoflurane between lipid bilayers and water, using racemic isoflurane and gas chromatography with a chiral column. For lipid bilayers of phosphatidylcholine (PC) and 4 mol% phosphatidic acid (PA), both with and without cholesterol (CHOL), we found equal partitioning of the isoflurane optical isomers. The ratios of the S(+) to R(-) isoflurane partition coefficients were (mean +/- SEM): 1.018 +/- 0.010 for bilayers of PC/CHOL/PA (mole ratios 56:40:4) and 1.011 +/- 0.002 for bilayers of PC/PA (mole ratio 96:4). Molar partition coefficients for racemic isoflurane were 49 +/- 4 and 165 +/- 10, respectively. These findings support the view that the stereoselective effects on ion channels observed with isoflurane are due to direct actions on proteins rather than lipids.

Effects of elevated pressures of inert gases on cytosolic free Ca2+ of cultured human neuroblastoma cells stimulated with carbachol: relevance to high pressure neurological syndrome

Suspended cells of the human neuroblastoma line SK-N-SH were exposed to elevated pressures of non-narcotic helium (He) and the narcotic gases nitrogen (N2), and argon (Ar) and stimulated with carbachol. He, 18 and 36 atmospheres absolute (ATA), equivalent to 544 and 1120 feet of seawater, potentiated the increase in [Ca2+]i induced by carbachol, as measured by Fura-2. Carbachol-stimulated increases in [Ca2+]i were not significantly altered from values in 1 ATA air by either N2 or Ar at the same pressures. The response to carbachol of cells exposed to 36 ATA of He and slowly decompressed to 1 ATA was indistinguishable from that of cells never exposed to pressure. Thus this pressure-potentiated increase in [Ca2+]i is compatible with excitation, is reversible and is not elicited by narcotic gases. It was observed, moreover, at pressures encountered by commercial deep-sea divers. The High Pressure Neurological Syndrome (HPNS) encountered by divers breathing He/O2 mixtures at high pressures, and its known antagonism by N2, may be due in part to effects on neuronal [Ca2+]i levels since an increase in these would most likely result in an excitatory response.

Effects of elevated pressures of inert gases on cytosolic free Ca2+ of human platelets stimulated with ADP

Platelets were exposed to elevated pressures of helium (He), nitrogen (N2), and He/xenon (He/Xe, 85/15), and stimulated with ADP (5 microM). He to 36 atmospheres absolute (ATA) inhibited the ADP-stimulated increase in [Ca2+]i, measured by Fura-2/AM, with most of the effect occurring by 9 ATA. N2 caused a similar but initially greater effect and it was bimodal, with inhibition being less at 36 ATA than at 18 ATA. N2 also significantly depressed basal levels at 18 and 36 ATA. In the He/Xe mixture, the effect on ADP-stimulated [Ca2+]i was no different from He alone at the same pressures, but basal levels were significantly depressed. In a calcium-free medium both He and N2 moderately depressed the already-reduced response to ADP stimulation but only N2 significantly depressed basal levels at all pressures. These results indicate that raised pressures of inert narcotic gases, as well as pressure per se (represented by He), can affect cell [Ca2+]i sufficiently to have physiological consequences. This, together with previous findings in this laboratory, may have implications for some of the neurological problems associated with deep diving.

Inhibition of protein kinase C by alcohols and anaesthetics

Despite almost a century of research, the mechanism of anaesthesia remains obscure and there is still no agreement on the location of the site(s) of action. Because the potencies of general anaesthetics increase in proportion to their solubility in olive oil, this led to a consensus that the site is within the cell membrane. This led to theories that lipid bilayer perturbation was the primary event, which was then transmitted to a membrane protein. But at the concentrations used clinically, such perturbations are small. A plausible site would be in or on ion channels at the synapse, where a number of modulatory effects have been described. A possible location for such a site would be at the protein-lipid interface. We report here that anaesthetics inhibit protein kinase C, a key component in signal transduction. The potency is a linear function of the octanol-water partition coefficient (the Meyer-Overton rule of anaesthesia). The effect was obtained in a lipid-free assay, implicating a hydrophobic site in the protein, supporting the contention that a (membrane) protein may be a target for anaesthetic interactions. In a lipid-dependent assay, a potential role of lipids in the protein-site model was demonstrated. The inhibition was absent in the isolated catalytic domain, suggesting that the site of inhibition is on the regulatory subunit, which is unique to protein kinase C.

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

The influence of local and general anaesthetics on cation influx through the fast, voltage-dependent sodium channel and the 5-HT3 receptor cation channel was studied in N1E-115 mouse neuroblastoma cells by measuring 2-min influx of the organic cation 14C-guanidinium induced by either veratridine (1 mmol/l) or 5-HT (100 mumol/l). The veratridine-induced influx of 14C-guanidinium was potentiated by scorpion toxin and inhibited by tetrodotoxin. The 5-HT-induced 14C-guanidinium influx was not affected by tetrodotoxin but it was inhibited by nanomolar concentrations of the selective 5-HT3 receptor antagonists ondansetron and ICS 205-930; at high micromolar concentrations these compounds also inhibited the veratridine-induced influx of 14C-guanidinium. The 14C-guanidinium influx through both channels was inhibited by local and general anaesthetics. The rank order of potency for inhibition of veratridine-induced influx by local anaesthetics was tetracaine > bupivacaine > cocaine > lidocaine > procaine and that for inhibition of the 5-HT3 receptor channel was tetracaine > bupivacaine > cocaine > procaine > lidocaine. With the exception of procaine and cocaine, which were equipotent at both channels, the local anaesthetics were 4.4-fold (lidocaine) to 25-fold (tetracaine) more potent at the fast sodium channel than at the 5-HT3 receptor channel. The rank order of potency for general anaesthetics was propofol > etomidate = alfaxalone = ketamine > thiopental = methohexital at the fast sodium channel, and propofol > or = etomidate > alfaxalone = methohexital > thiopental > ketamine at the 5-HT3 receptor channel.(ABSTRACT TRUNCATED AT 250 WORDS)

Molecular interactions of anaesthetics with biological membranes

1. There is not yet a consensus as to which of the neuronal membranes, which molecular component of any particular membrane or what specific function of the membrane is critical for general anaesthesia. 2. However, when considering anaesthetic effects on different synapses, with neurotransmitter receptors, ion channels etc., the unifying central concept is action at a membrane level. 3. This paper will review the general evidence for this unifying hypothesis, and consider the apparent exceptions and limitations. 4. The membrane hypothesis is usually stated in the form of the Meyer and Overton "rule" relating anaesthetic potency to hydrophobic solubility. 5. The relationship applies to inhaled anaesthetics with potencies over a 100,000-fold range and has been described as one of the most powerful correlations in biology. 6. Finding additional compounds that confirm this correlation is not likely to elucidate further the anaesthetic mode of action, and concentrating on the apparent exceptions to the hypothesis may prove to be a better approach. 7. The apparent exceptions to the membrane hypothesis include some of the physiological and convulsant gases as well as higher members of a homologous series of hydrophobic compounds above the so-called "cut-off" effect. 8. The apparent limitations include two of the most widely used clinical agents--enflurane and isoflurane. 9. This paper will include some new data characterizing the anaesthetic site of action using a diverse group of anaesthetics (including some of the exceptions already mentioned). 10. The evidence is against an earlier hypothesis that there is a bimodal distribution of the molecular sites.

Effects of temperature on the anaesthetic potency of halothane, enflurane and ethanol in Daphnia magna (Cladocera: Crustacea)

1. The effects of temperature on the anesthetic potencies of halothane, enflurane and ethanol have been studied in the water flea Daphnia magna. 2. In the absence of anaesthetics, decreasing temperature resulted in decreased activity by the daphnids. 3. Potencies in the gas phase decreased with increasing temperature for all of the anaesthetics, while aqueous potency decreased for halothane and enflurane but increased for ethanol. 4. Enthalpy calculations suggest that the observed potency changes for the inhalational anaesthetics cannot be accounted for in terms of changing solubility in lipid bilayers but most likely reflect more specific interactions with animal target sites.

Effects of pressure and gaseous anesthetics on the aggregation of human blood platelets and marine sponge cells: similarities in responses

1. Aggregation of blood platelets and marine sponge cells was inhibited by nitrous oxide (N2O) and helium (He) at elevated pressure but potentiated by xenon (Xe) despite the fact that Xe and N2O are equipotent gaseous anesthetics. 2. The above aggregations are dependent on an extracellular source of calcium. 3. Platelet aggregation induced by phorbol myristate is independent of extracellular calcium and was inhibited by both N2O and Xe and by high pressures of He. 4. The results argue against a common mechanism for Xe and N2O and suggest that pressure may affect calcium interactions with binding site.

Anesthetic inhibition of firefly luciferase, a protein model for general anesthesia, does not exhibit pressure reversal

The surprising observation that pressures of the order of 150 atmospheres can restore consciousness to an anesthetized animal has long been central to theories of the molecular mechanisms underlying general anesthesia. We have constructed a high-pressure gas chamber to test for "pressure reversal" of the best available protein model of general anesthetic target sites: the pure enzyme firefly luciferase, which accounts extremely well for animal potencies (over a 100,000-fold range). We found no significant pressure reversal for a variety of anesthetics of differing size and polarity. It thus appears that either firefly luciferase is not an adequate model for general anesthetic target sites or that pressure and anesthetics act at different molecular sites in the central nervous system.

Stereospecific effects of inhalational general anesthetic optical isomers on nerve ion channels

Although it is generally agreed that general anesthetics ultimately act on neuronal ion channels, there is considerable controversy over whether this occurs by direct binding to protein or secondarily by nonspecific perturbation of lipids. Very pure optical isomers of the inhalational general anesthetic isoflurane exhibited clear stereoselectivity in their effects on particularly sensitive ion channels in identified molluscan central nervous system neurons. At the human median effect dose (ED50) for general anesthesia, the (+)-isomer was about twofold more effective than the (-)-isomer both in eliciting the anesthetic-activated potassium current IK(An) and in inhibiting a current mediated by neuronal nicotinic acetylcholine receptors. For inhibiting the much less sensitive transient potassium current IA, the (-)-isomer was marginally more potent than the (+)-isomer. Both isomers were equally effective at disrupting lipid bilayers.

Saturable binding of anesthetics to nicotinic acetylcholine receptors. A possible mechanism of anesthetic action

Recent controversies in the existence of saturable binding of general anesthetics in brain tissues prompted a careful examination of specific binding of anesthetics to neural receptors. We examined the binding of both local and general anesthetics using electron spin resonance and radioligand criteria. Our results suggested that the hydrophobic path, most probably through the lipid bilayer, figures importantly in the binding of the uncharged moieties of anesthetics. Competitive interactions by hydrophobic compounds for the high-affinity site in the nicotinic acetylcholine receptor led us to propose a hypothesis that includes a hydrophobic crevice of limited volume as part of the high-affinity site. Association of anesthetic at this crevice is in turn dependent on the anesthetic concentration in the lipid phase of the membrane. The hypothesis provides a mechanism for the saturable interaction of anesthetics with their protein target site in the membrane without violating the correlations expressed by the Meyer-Overton rule of anesthetic action.

Inhalational anesthetics stereochemistry: optical resolution of halothane, enflurane, and isoflurane

Halothane (I), enflurane (II), and isoflurane (III), which are among the most important inhalation anesthetics, are currently administered as racemic mixtures. The pure enantiomers have not been described, and no analytical method for resolving the commercially available racemic mixtures has been reported. Complete optical resolution of (+/-)-I and (+/-)-III on per-n-pentylated alpha-cyclodextrin (Lipodex A) and of (+/-)-II and (+/-)-III on octakis(6-O-methyl-2,3-di-O-pentyl)-gamma-cyclodextrin capillary columns has been achieved, making rapid and convenient determination of enantiomeric ratios in samples of all three of these anesthetics possible.

Probing the molecular dimensions of general anaesthetic target sites in tadpoles (Xenopus laevis) and model systems using cycloalcohols

1. The series of cycloalcohols C6, C7, C8 and C10 have been used to probe the molecular dimensions of a variety of general anaesthetic target sites. 2. The general anaesthetic EC50 concentrations of the cycloalcohols were determined for tadpoles (Xenopus laevis). All of the cycloalcohols tested were found to be potent general anaesthetics (on average EC50/Csat = 0.03). 3. The effects of the cycloalcohols on highly purified luciferase enzymes from fireflies (Photinus pyralis) and bacteria (Vibrio harveyi) were also investigated. Both enzymes were inhibited competitively, with the cycloalcohols competing with firefly luciferin for binding to the firefly enzyme and with n-decanal for binding to the bacterial enzyme. 4. The binding site on the firefly enzyme could accommodate two molecules of cycloalcohols C6 and C7 but only a single molecule of the larger cycloalcohols (C8 and C10), implying a volume of the binding site of about 250 cm3 mol-1. In contrast, the binding site on the bacterial luciferase could bind only a single cycloalcohol molecule between C6 and C10. 5. While all of the cycloalcohols were potent inhibitors of the firefly luciferase enzyme (on average EC50/Csat = 0.015), they were very weak inhibitors of the bacterial luciferase enzyme (on average EC50/Csat = 0.12). Since both enzymes bind long-chain aliphatic n-alcohols tightly, the differing affinities of the cycloalcohols for the two enzymes is probably a consequence of geometrical factors. 6. The cycloalcohols produced very small effects on lipid bilayers. At EC50 concentrations which produce general anaesthesia, lipid bilayer phase transitions were shifted, on average, by only 0.43 degrees C. 7. We conclude that the general anaesthetic effects of the cycloalcohols can most economically be explained by assuming that the cycloalcohols act at protein binding sites in the central nervous system. These target sites would have binding properties similar to those of the anaesthetic-binding site on firefly luciferase, but their average volume would be somewhat smaller than 250 cm3 mol -1.

Inhibition of choline transport in erythrocytes by n-alkanols

The choline transport system of erythrocytes is reversibly inhibited by ethanol, n-butanol, n-hexanol, n-octanol, and n-decanol, but not by n-dodecanol. Each methylene group in the alkyl chain contributes 560 cal/mol to the free energy of binding at the inhibitory site. Inhibition results from the cooperative binding of two molecules of an alcohol, judging by the Hill coefficient n of 1.7-1.9. The mechanism of inhibition is noncompetitive, and the partition of the carrier between inward-facing and outward-facing forms is unaffected by the alcohols; it follows that the four main carrier forms, the inner and outer free carrier, and the inner and outer carrier-substrate complex, are equally susceptible to inhibition. Hexanol and decanol accelerate the reaction of N-ethylmaleimide with a thiol group in the inner carrier channel, but ethanol and butanol, at concentrations that inhibit transport by 70%, do not. The disproportionate effects on substrate transport and the N-ethylmaleimide reaction are most simply explained as the direct result of binding of alcohol molecules in different regions of the carrier, rather than as the indirect result of a disturbance in the structure of the lipid bilayer induced by the alcohols.

The effect of n-alcohols on vesicular permeability induced at the lipid phase transition temperature: a 1H-NMR study

The effect of a series of n-alcohols on the permeability of small, unilamellar dipalmitoyl phosphatidylcholine (DPPC), dimyristoyl phosphatidylcholine (DMPC) and distearoyl phosphatidylcholine (DSPC) vesicles at the gel-to-liquid crystal phase transition temperature was investigated. It was found that the permeability took the form of the transient lysis of a fraction of the population of vesicles. The effect on this lysis of the n-alcohols was seen to be very chain-length dependent, with a minimum at n = 8 (octan-1-ol) for DPPC vesicles. A similar minimum was observed in the presence of 0.1 mM Triton X-100, but the detergent could then interact with certain of the alcohols to produce permanent channels. The results are discussed in terms of the semi-empirical model of Brasseur et al. (1985) Biochim. Biophys. Acta 814, 227-236, for the interaction of the n-alcohols with a DPPC membrane. The effect of various n-alcohols on the outer and inner monolayers of DPPC vesicles was also studied and the results related to their fluidising effect, allowing channels to open at the phase transition temperature.

Volatile general anaesthetics activate a novel neuronal K+ current

Although it is still controversial whether the primary target sites underlying general anaesthesia are proteins or lipids, it is generally thought that the ultimate targets are ion channels in nerve membranes. One approach to finding these targets is to study the effects of general anaesthetics on identified neurons, where differential effects on neuronal activity can be pursued to the molecular level. Here we report that amongst a group of apparently identical molluscan neurons having endogenous firing activity, a single cell displays an unusual sensitivity to volatile agents (which, at surgical levels, completely inhibit its activity). We further show that this sensitivity is due to a novel anaesthetic-activated K+ current, which is found in the sensitive cell but not in the surrounding insensitive cells. This K+ conductance is not appreciably voltage-gated and persists for as long as the anaesthetic is present. The response to anaesthetics is completely reversible and saturates at low anaesthetic partial pressures: the half-maximal response for halothane occurs at 0.0063 atm, close to its minimum alveolar concentration (0.0075 atm) in man.

The effects of inhalation anesthetics on calcium-stimulated exocytosis in a natural membrane model system

Sea urchin egg cortices were used as an in vitro natural membrane model system to determine the effects of inhalation anesthetics on the Ca2+-regulated exocytotic fusion of cortical vesicles with the egg plasma membrane. When Ca2+ was either absent or present in amounts below the threshold for exocytosis, methoxyflurane, halothane, enflurane, isoflurane, chloroform and fluoroxene, at concentrations up to 5 mM, had no effect on the fusion of cortical vesicles with the plasma membrane. However, when Ca2+ was present at or above threshold levels for exocytosis, each of the tested anesthetics caused an inhibition of cortical vesicle fusion. Exocytosis was inhibited most effectively by methoxyflurane (55%), followed by halothane (30%), while fluoroxene consistently had the least effect (less than 5%). These observations support the view that volatile anesthetics can impair the Ca2+-regulated fusogenic activities of natural membranes and are consistent with other data showing that inhalational agents inhibit secretory processes in intact cells.

Phospholipid/cholesterol membranes containing n-alkanols: a 2H-NMR study

The influences of 1-octanol and 1-decanol on aqueous multilamellar dispersions of 1-hexadecanoyl(octadecanoyl)-2-[2H31]hexadecanoyl-sn-glycero -3-phosphorylcholine (PC-d31)/cholesterol (3:1) have been examined using 2H-NMR. The gel to liquid crystalline phase transition of the PC-d31/cholesterol dispersion is modulated by the addition of 1-alkanol, which reduces the onset temperature and increases the width of the transition. 1-Octanol has a greater effect on the transition onset and completion temperatures than does 1-decanol, as determined from analysis of the temperature-dependent 2H-NMR spectra. 2H-NMR C-2H bond order parameters as a function of phospholipid acyl chain position at 60 degrees C, where all dispersions are fully liquid crystalline, have been calculated from the depaked spectra. 1-Decanol reduces the phospholipid order by only 2%. This can be attributed to the lower effective cholesterol concentration in the 1-alkanol/PC-d31/cholesterol dispersions. 1-Octanol, however, reduces the phospholipid order by 10% at 60 degrees C. Correlations between the effects of 1-octanol and 1-decanol on phospholipid order parameters and phospholipid/cholesterol phase transitions are discussed.

Pressure reversal of alphaxalone/alphadolone and methohexitone in tadpoles: evidence for different molecular sites for general anaesthesia

Tadpoles were used to study quantitative interactions between high pressure and two intravenous anaesthetics, alphaxalone/alphadolone and methohexitone. The potencies of the two agents were decreased by high pressure but to different extents. The maximum effect was seen in the pressure range 70-130 atmospheres absolute (ATA). The increases in the normobaric anaesthetizing concentration (ED50) required at 100 ATA were alphaxalone/alphadolone:405 +/- 5 (s.d.)%; methohexitone:658 +/- 80 (s.d.)%. For both alphaxalone/alphadalone and methohexitone, the curves obtained when the increase in ED50 was plotted against increasing pressure showed plateaux at pressures above 70 ATA. These data support the concept of the two intravenous drugs causing general anaesthesia by the occupation of separate molecular 'sites' with different but finite capacities.

Halothane-induced changes in acetylcholine receptor channel kinetics are attenuated by cholesterol

The single-channel recording technique was used to investigate the role of membrane lipids in the action of general anesthetics on ion channels. We examined the effects of halothane on acetylcholine receptor channels in Xenopus laevis myocytes in which the plasma membrane cholesterol level had been changed by pretreatment with cholesterol-rich or cholesterol-free liposomes. We found that the alteration in acetylcholine receptor channel kinetics, elicited in the presence of clinically-relevant concentrations of halothane, is attenuated when membrane cholesterol is increased and enhanced when membrane cholesterol concentration is decreased. These findings support the idea that general anesthetics interact with synaptic receptor channels indirectly through the lipid domains in which these synaptic proteins are embedded.

A deuterium NMR study of labelled n-alkanol anesthetics in a model membrane

The 2H-NMR spectra of 50 wt.% aqueous multilamellar dispersions of dipalmitoylphosphatidylcholine (DPPC) containing either selectively deuterated 1-decanol (25 mol%) or [2H17]-1-octanol (25 mol%) have been measured as a function of temperature. Both alkanols are potent anesthetics. A detailed carbon-deuterium bond order parameter profile of 1-decanol in liquid crystalline phospholipid dispersions at 50 degrees C was determined from the quadrupolar splittings of 1-decanols deuterated at eight different positions. A maximum order parameter SCD = 0.20 was obtained for [5,5-2H2]-1-decanol, with labels at both ends of the 1-decanol exhibiting reduced order parameters. Explanations for the reduced order towards the hydroxyl group of 1-decanol are discussed in terms of either increased amplitudes of motion or geometric effects due to hydrogen bonding. By comparing the order parameter profile of sn-2 chain deuterated phosphatidylcholine dispersions containing 25 mol% 1-decanol (J.L. Thewalt, S.R. Wassall, H. Gorrissen and R.J. Cushley, Biochim. Biophys. Acta, 817 (1985) 355) with the profile of deuterated 1-decanol in DPPC, we estimate that decanol is approximately parallel to the C-3 to C-13 region of the phosphatidylcholine's sn-2 chain. Variation of the spectral moments M1 with temperature indicates that both 1-decanol and 1-octanol are sensitive to the packing of the lipid in which they are dissolved. Below the phase transition temperature, the 2H-NMR spectra of either 1-decanol (selectively deuterated) or 1-octanol (perdeuterated) are broad powder patterns, characteristic of axially symmetric rotation about the alcohol's long axis. This is in contrast to the 2H-NMR spectra obtained from deuterated phosphatidylcholine under similar conditions, which implies that the phospholipid acyl chain conformations are more restricted than those of the alcohol at these temperatures. From the M1 behavior of the various alkanol chain segments with temperature, the gel to liquid crystalline phase transition is seen to initiate in the middle of the DPPC/1-alkanol bilayer.

General and local anaesthetics perturb the fusion of phospholipid vesicles

The effects of general and local anaesthetics on Ca2+-induced fusion of negatively charged lipid vesicles have been investigated. Vesicles composed of phosphatidylcholine and phosphatidic acid (2:1 molar ratio) were induced to fuse using 5 mM free Ca2+. Fusion, assessed by an increase in size using gel filtration techniques and confirmed by electron microscopy, displayed a dependence on Ca2+ and Mg2+ concentration and on temperature. The inhalational anaesthetics halothane, methoxyflurane and diethyl ether enhanced fusion as did the uncharged local anaesthetic benzocaine. In contrast, the charged local anaesthetics lignocaine and bupivacaine inhibited the fusion process. It is suggested that the enhancement observed with the inhalational anaesthetics and benzocaine was mediated by an effect on lipid fluidity and the inhibition observed with the charged tertiary amine anaesthetics was due to an antagonism towards Ca2+.