General Anesthetics Minimally affect Lipid Bilayer Properties at Clinical Concentrations

Abstract

General anesthetics are widely used but their exact molecular mechanisms are incompletely understood. One early model, known as the unitary lipid-based hypothesis of anesthetic action, proposed that nonspecific anesthetic interactions with the lipid bilayer altered neuronal function through effects on membrane properties. This model, which is based largely on the Meyer-Overton correlation between anesthetic potency and lipid solubility, has been challenged by more recent evidence supporting direct interactions of anesthetics with a range of proteins, in particular transmembrane ion channels. The available results, however, do not exclude anesthetic modulation of ion channel function through interactions with the lipid bilayer. We therefore tested the effects of a range of chemically diverse general anesthetics (diethyl ether, desflurane, sevoflurane, isoflurane, halothane, chloroform, bromoform, cyclopropane, F3, ketamine, etomidate and propofol) and related nonanesthetics (F6 and flurothyl, predicted to be anesthetic based on their lipid solubilities) on lipid bilayer properties using a sensitive gramicidin-based assay. We show that none of the anesthetics or nonanesthetics tested significantly altered lipid bilayer properties at clinically relevant concentrations. Even at supra-anesthetic concentrations, most anesthetics produced only minimal bilayer-mediated effects, though some were able to alter lipid bilayer properties at supra-therapeutic (toxic) concentrations. We conclude that the effects of general anesthetics at clinically relevant concentrations involve minimal changes in lipid bilayer properties. Therefore clinical anesthesia is likely to result from direct interactions of anesthetics with membrane proteins.Supported by GM058055 (HCH) and GM021347 (OSA).