A quantitative description of QX222 blockade of sodium channels in squid axons

The interaction of QX222, a quaternary ammonium derivative of lidocaine, with the Na channel was studied in internally perfused squid axons under voltage-clamped conditions. A use-dependent block was observed in response to repetitive depolarizing pulses. The time constant for block development and the steady state level of the block were increased with increasing frequency of stimulation from 0.1 to 10 Hz. Use-dependent block can be viewed as a net increase in the drug incorporation into Na channels with successive pulses. That is, net drug uptake by Na channels occurs during the depolarizing phase and net drug release occurs during the interpulse interval. The observed uptake rate of use-dependent block is shown to be a linear combination of the uptake rates associated with the depolarizing and resting potentials. Also, the steady state fraction of blocked channels is shown to be a linear combination of the state-dependent blockade equilibria. Drug-channel interactions are assumed to be dependent on gated control of the diffusion path between drug pool and the interior channel binding site. Drug ingress to the binding site can be inhibited by the channel gates (receptor guarding), while drug bound to the channel may become trapped by closure of the channel gates (trapping). On the basis of these assumptions, a simple procedure is proposed for estimating apparent rate constants governing the drug-channel binding reactions for two cases of channel blockade. The estimated forward (k) and backward (1) rate constants are: 2.45 x I05 M-1 s- and 0.23 x 103 s-1, respectively, for k and I for the case when the drug is trapped by both activation and inactivation gates, and 3.58 x 105 M-l s-l and 4.15 x 10-3 S-l for the case when the drug is not trapped. While these two schemes make a similar prediction with respect to the resulting uptake rates, their prediction of the steady state level of block differs. The observed steady state level of block could quantitatively be predicted by the trapped scheme but not by the untrapped scheme, thus providing a means for discriminating between these two schemes. In addition, the trapped scheme, but not the untrapped scheme, could provide an explanation for the observed voltage dependence of the slow recovery from use-dependent block.