BACKGROUND Despite their key role in the generation and propagation of action potentials in excitable cells, voltage-gated sodium (Na+) channels have been considered to be insensitive to general anesthetics. The authors tested the sensitivity of neuronal Na+ channels to structurally similar anesthetic (1-chloro-1,2,2-trifluorocyclobutane; F3) and nonanesthetic (1,2-dichlorohexafluorocyclobutane; F6) polyhalogenated cyclobutanes by neurochemical and electrophysiologic methods. METHODS Synaptosomes (pinched-off nerve terminals) from adult rat cerebral cortex were used to determine the effects of F3 and F6 on 4-aminopyridine- or veratridine-evoked (Na+ channel-dependent) glutamate release (using an enzyme-coupled spectrofluorimetric assay) and increases in intracellular Ca2+ ([Ca2+]i) (using ion-specific spectrofluorimetry). Effects of F3 and F6 on Na+ currents were evaluated directly in rat lumbar dorsal root ganglion neurons by whole-cell patch-clamp recording. RESULTS F3 inhibited glutamate release evoked by 4-aminopyridine (inhibitory concentration of 50% [IC50] = 0.77 mM [approximately 0.8 minimum alveolar concentration (MAC)] or veratridine (IC50 = 0.42 mM [approximately 0.4 MAC]), and veratridine-evoked increases in [Ca2+]i (IC50 = 0.5 mM [approximately 0.5 MAC]) in synaptosomes; F6 had no significant effects up to 0.05 mM (approximately twice the predicted MAC). F3 caused reversible membrane potential-independent inhibition of peak Na+ currents (70+/-9% block at 0.6 mM [approximately 0.6 MAC]), and a hyperpolarizing shift in the voltage-dependence of steady state inactivation in dorsal root ganglion neurons (-21+/-9.3 mV at 0.6 mM). F6 inhibited peak Na+ currents to a lesser extent (16+/-2% block at 0.018 mM [predicted MAC]) and had minimal effects on steady state inactivation. CONCLUSIONS The anesthetic cyclobutane F3 significantly inhibited Na+ channel-mediated glutamate release and increases in [Ca2+]i. In contrast, the nonanesthetic cyclobutane F6 had no significant effects at predicted anesthetic concentrations. F3 inhibited dorsal root ganglion neuron Na+ channels with a potency and by mechanisms similar to those of conventional volatile anesthetics; F6 was less effective and did not produce voltage-dependent block. This concordance between anesthetic activity and Na+ channel inhibition supports a role for presynaptic Na+ channels as targets for general anesthetic effects and suggests that shifting the voltage-dependence of Na+ channel inactivation is an important property of volatile anesthetic compounds.