UNIDIRECTIONAL SODIUM AND POTASSIUM FLUXES THROUGH THE SODIUM CHANNEL OF SQUID GIANT AXONS DAVID BUSATH AND TED BEGENISICH Department of Physiology, University of Rochester School of Medicine, Rochester, New York 14642 ABSTRACT Unidirectional 22Na-traced sodium influx or 42K-traced potassium efflux across the membranes of voltage-clamped squid giant axons was measured at various membrane potentials under bi-ionic conditions. Tetrodo- toxin almost entirely eliminated the extra K+ efflux induced by short repetitive depolarizations in the presence of tetraethylammonium or 3,4-diaminopyridine. A method of determining the voltage dependence of the unidirectional flux through voltage-gated channels is described. This technique was used to obtain the unidirectional flux-voltage relation for the sodium channel in bi-ionic and single-ion conditions. It allows the determination of the unidirectional flux at the zero-current potential which, for influx, was found to be -20% of the value measured 80 mV negative to the zero-current potential. The unidirectional flux ratio under bi-ionic conditions was also measured and the flux ratio exponent found to average 1.15 with an external sodium solution and an internal potassium solution. A three-barrier, two-site, multi-occupancy model previously obtained for other conditions was found to predict a similar non-unity average for the flux ratio exponent. It is also shown that some single-occupancy models can predict non-unity values for the flux ratio exponent in bi-ionic conditions. INTRODUCTION Most studies of the ionic permeation processes of the sodium and potassium channels in nerve have employed measurements of ionic currents and zero current potentials (Hille, 1975; Begenisich and Cahalan, 1980 a, b). The results of these studies have shown that the sodium channel of squid giant axons behaves under some conditions like a multi-ion pore. A particular permeation model has been presented which can account for most of the electrical measurements (Begenisich and Cahalan, 1980 a, b). It is possible to study the unidirectional ionic fluxes as well as currents in giant axons. We have already reported measurements of the ratio of the unidirectional influxes and effluxes of sodium ions through the sodium channel of these nerve fibers (Begenisich and Busath, 1981). In the present study we measure 22Na influx and 42K efflux in fibers bathed in K-free seawater and perfused with Na- free internal solutions. We describe a method of deter- mining the voltage dependence of the individual unidirec- tional ionic fluxes. We apply this technique to axons with both K-free and Na-free internal solutions. We also deter- mine the ratio of unidirectional K efflux and Na influx and discuss the significance of the measurement of flux ratios under bi-ionic conditions. Lastly, the predictions of the fluxes from the three-barrier, two-site model determined Dr. Busath's present address is Department of Physiology and Biophysics, University of Texas Medical Branch, Galveston, TX 77550. BIOPHYS. J. Biophysical Society * 0006-3495/82/10/041/00' Volume 40 October 1982 41-49 from electrical measurements are compared with the pres- ent results. METHODS The methods used in this study have been described previously (Begeni- sich and Busath, 1981; Begenisich and De Weer, 1980). These experi- ments were performed at IOOC on well-cleaned giant axons of Loligo pealei obtained from the Arrive Alive Biological Supply Company, Long Island, NY. Both live and decapitated animals were used. The diameters of the 25 axons used in this study ranged from 400 to 700 gm, with an average near 500 Am. The resting potential of axons bathed in artificial seawater and internally perfused with KSIS (potassium standard internal solution, see below) averaged -63 mV. The internal perfusion and voltage-clamp techniques were similar to those used by Begenisich and Lynch (1974). Levis (1981) has described a method for improving series resistance compensation by matching the phases of the membrane voltage and series resistance compensation signals. We have used a simplified version of this technique. An additional benefit of this improvement is a decrease in clamp settling time. Fig. I shows the capacitative transient from a typical axon produced by a - 30-mV pulse. The transient is essentially complete within 10 ,us. In several axons this could be reduced to 8 us. The decrease in settling time and the improved series resistance compensation provide a more reliable measure of the tail currents used to determine the instanta- neous current-voltage relations described below. The series resistance compensation used ranged from 2 to 6 (1. cm2 but was usually 4-5 i2. cm2 with an average value of 4.3 (1. cm2. This average value compares favorably with other measurements of the total (preparation plus elec- trode) measured series resistance of 3.03 Q-cm2 (Binstock et al., 1975) and 4.02 j2.cm2 (Salzberg et al., 1981; B. Salzberg, personal communica- tion). Membrane voltages have been corrected for the junction potential between the 0.56 M KCI internal electrode and the internal perfusion 9 $1.00 41