Zinc has been shown to modulate hemichannel currents of connexins Cx35 and Cx38 in Xenopus oocytes (1).In both cases the effects were biphasic; i.e., low concentrations of zinc enhanced, whereas higher concentrations decreased, the magnitudes of the voltage-activated hemichannel currents. The present study was designed to determine the effects of zinc on hemichannels formed by Cx26, a connexin reportedly expressed on dendrites of carp horizontal cells and implicated in a mechanism for photoreceptor feedback (2,3,4). In addition, we examined whether histidine, a zinc chelator, would block the action of zinc on Cx26 hemichannel currents, or would exert a direct effect on those currents.Methods for oocyte preparation and recording of connexin hemichannel currents followed procedures described previously (5). Briefly, Stage V-VI oocytes were removed from gravid female Xenopus, enzymatically dissociated and defolliculated, and then used to express human Cx26 connexin in the presence of an antisense oligomer to the endogenous oocyte connexin (Cx38 Voltage-activated membrane currents were recorded with a two-electrode voltage clamp. The two protocols we used gave equivalent results: (1) with the cell clamped at 0 mV, 10-s voltage steps were imposed from Ϫ50 mV to ϩ50 mV in 10-mV increments, and (2) from the holding potential of 0 mV, currents were recorded in response to voltage ramps extending from Ϫ100 mV to ϩ60 mV at a rate of 0.04 mV/ms. Figure 1 compares a representative record from an oocyte expressing Cx26 with those obtained from an antisenseinjected cell and one that was not injected but expressed Cx38, its endogenous connexin. Note the efficacy with which the antisense oligo suppressed the current mediated by Cx38, and the significantly greater membrane currents of Cx26 in response to both depolarizing and hyperpolarizing voltages. Figure 2A shows current recordings from oocytes expressing human Cx26 in MB and after the addition of various concentrations of zinc. In MB, a voltage ramp from Ϫ100 mV to ϩ60 mV elicited large currents at all potentials positive and negative to the reversal potential (Ϫ10 mV), suggesting that the hemichannels are constitutively open (5). In the presence of low (1 M) concentrations of zinc, the magnitude of the currents through Cx26 hemichannels was substantially enhanced. Higher concentrations of zinc, however, decreased these currents in a dose-dependendant fashion such that 10 M zinc reduced the hemichannel currents to the level seen in control MB solution, and 100 M and 1 mM zinc further decreased the hemichannel currents. The bar graphs of Figure 2B, taken from the data in Figure 2A, illustrate the effects of zinc on Cx26 hemichannel currents recorded at ϩ40 mV. Both the enhancement and inhibition by zinc of the Cx26 hemichannel current showed little voltage dependence. As shown in Figure 2C, the ratio of the membrane currents measured in zinc to those recorded in MB were virtually unchanged across the range of membrane voltage tested (from Ϫ100 mV to ϩ60 mV) for both low conc...