Local anaesthetic block of sodium channels: raising the barrier

The mechanism by which local anaesthetics block voltage-gated sodium channels has long been an intriguing puzzle for electrophysiologists. The effect of these molecules depends upon membrane potential, channel conformation, pH and access to the ion-conducting pore. A central tenet has been that local anaesthetics bind to a site in the lumen of the channel and physically occlude it, thus preventing the flow of current. This idea has been challenged with the advent of molecular models of the depolarized (open/inactivated) sodium channel pore (Lipkind & Fozzard, 2005) based on X-ray structures of potassium channels (Jiang et al. 2004). Docking local anaesthetic molecules in these open pores at binding sites that had been identified previously by site-directed mutagenesis (Ragsdale et al. 1994) resulted in the surprising conclusion that many drug molecules were too small to fully occlude the pore (Lipkind & Fozzard, 2005). This led to the alternative hypothesis that local anaesthetics may actually prevent current flow through sodium channels by introducing a positive charge that electrostatically impedes the flow of sodium ions, rather than by physically blocking them. In a study published in this issue of The Journal of PhysiologyMcNulty et al. (2007) have combined experimental and modelling approaches to test and further develop this hypothesis.

The position of the binding site for lidocaine and other local anaesthetics was defined in mutagenesis experiments. Those experiments identified a key phenylalanine (F1759 in the human Na_V1.5 channel studied here) that was required for use and voltage-dependent block by local anaesthetics (Ragsdale et al. 1994). This phenylalanine lies in transmembrane α helix IVS6, which lines the intracellular portion of the channel pore. Most local anaesthetics Contain a tertiary amine that is predominantly charged at physiological pH. This charge has been proposed to bind to F1579. Hydrophobic substituents at the other end of local anaesthetic molecules have been proposed to bind a tyrosine residue (Y1766 in human Na_V1.5) located two helical turns more intracellularly in IVS6. These binding determinants align the drug molecule along the wall of the channel with its charged tertiary amine group just intracellular to the selectivity filter of the channel pore.

To test the influence of charge at this position in the pore experimentally, McNulty et al. (2007) introduced charge at this position by replacing F1759 with charged amino acids and examined effects on single channel conductance, a parameter that is independent of channel expression or gating properties. Introduction of positively charged lysine or arginine reduced single channel conductance. Thus, introducing positive charge at this level in the channel pore can reduce conductance. However, introduction of positive charge per se did not completely block sodium current.

Would positive charge at this same position, but in the context of a local anaesthetic molecule, have a greater effect? By combining a molecular model of the sodium channel pore with calculations of the intrapore electrical potential profile, the authors compared the effects of positive charge at this position when provided as an amino acid side chain (analogous to the experiments) to the effects of a charge in the context of a local anaesthetic molecule docked in the ion channel pore. The model of the native channel predicted that negatively charged amino acids at the selectivity filter of the pore and in a ring of negatively charged amino acids just extracellular to the selectivity filter produce a strong negative electrical potential that extends through the selectivity filter and into the intracellular portion of the pore. This negative field would favour conduction of sodium ions. Substituting positively charged amino acids for F1759 in the model made the predicted potential just intracellular to the selectivity filter more positive, producing a barrier to the passage of positively charged ions. However, a local anaesthetic molecule bound in the pore with its positive charge near F1759 was calculated to raise the energy barrier for a sodium ion traversing the pore by almost twice as much as the positively charged amino acids. Whereas the side chains of the substituted amino acids were surrounded by water, the local anaesthetic filled a substantial volume and provided a hydrophobic, low dielectric environment for its positive charge. This in turn increased the positive potential associated with the charge and the distance over which that charge was felt. The drastically increased energy barrier produced by the local anaesthetic, in combination with any steric hindrance to ion movement that it produces, provides the basis for full block of the current.

The experimental results of this study provide an important proof of the principle that positive charge at the local anaesthetic binding site can indeed reduce current through the sodium channel pore. The modelling complements these results and proposes a mechanism for why charged local anaesthetics might be particularly effective in producing electrostatic block. Necessarily, such models are based on many assumptions. Key among these is the model of the channel pore (based on a symmetrical potassium channel rather than an asymmetrical sodium channel) and the position of the local anaesthetic within it. The calculations of electrical potential profile are also sensitive to many factors that are largely unknown. Despite these caveats, the present results provide a first experimental test of a new paradigm for local anaesthetic block. Future experiments in combination with refined models based on increasingly accurate structural information will define the relative roles of electrostatic repulsion and pore occlusion as modes by which local anaesthetics and other compounds prevent flow of current through sodium channels.