Ionic mechanism of isoflurane's actions on thalamocortical neurons.

We studied the actions of isoflurane (IFL) applied in aqueous solutions on ventrobasal neurons from thalamic brain slices of juvenile rats. By using the whole cell, patch-clamp method with current- and voltage-clamp recording techniques, we found that IFL increased a noninactivating membrane conductance in a concentration-dependent reversible manner. In an eightfold concentration range that extended into equivalent in vivo lethal concentrations, IFL did not produce a maximal effect on the conductance; this is consistent with a nonreceptor-mediated mechanism of action. TTX eliminated action potential activity but did not alter IFL effects. The effects on the membrane potential and current induced by IFL were voltage independent but depended on the external [K+], reversing near the equilibrium potential for K+. External Ba2+ or internal Cs+ applications, which block K+ channels, suppressed the conductance increase caused by IFL. External applications of the Ca2+ channel blockers Co2+ or Cd2+ or internal application of the Ca2+ chelator 1,2-bis-(2-aminophenoxy)-ethane-N,N, N',N'-tetraacetic acid did not prevent the effects of IFL, implying little involvement of Ca2+-dependent K+ currents. A contribution of inwardly rectifying K+ channels to the increased steady-state conductance seemed unlikely because IFL decreased inward rectification. An involvement of ATP-mediated K+ channels also was unlikely because application of the ATP-mediated K+ channel blocker glibenclamide (1-80 microM) did not prevent IFL's actions. In contrast to spiking cells, IFL depolarized presumed glial cells, consistent with an efflux of K+ from thalamocortical neurons. The results imply that a leak K+ channel mediated the IFL-induced increase in postsynaptic membrane conductance in thalamic relay neurons. Thus a single nonreceptor-mediated mechanism of IFL action was responsible for the hyperpolarization and conductance shunt of voltage-dependent Na+ and Ca2+ spikes, as reported in the preceding paper. Although anesthetics influence various neurological systems, an enhanced K+ leak generalized in thalamocortical neurons alone could account for anesthesia in vivo.