Fig. 3. Voltage-gating properties of homotypic and heteromeric Cx46 channels. (A-C) I_j values recorded from oocyte pairs were plotted as a function of time to reveal that channels expressing (A) wild-type Cx46 alone and mixed channels containing (B) both Cx46 and Cx50-S50P proteins as well as channels expressing (C) wild-type Cx46 and Cx50 subunits displayed altered voltage-gating sensitivity as co-injected pairs appear more responsive at larger voltage applications. Analysis of channel closure kinetics. The initial 250 mseconds of junctional current was fit to a monoexponential decay model to determine the mean time constant . (E,D) Representative I_j decays show that during an +80 mV potential channels containing (E) both Cx46 and Cx50-S50P subunits closed in 101 mseconds a value significantly faster then the 132 mseconds mean value exhibited by the (D) homotypic Cx46 channel (P>0.05, ANOVA). (F) Interestingly, channels co-expressing both wild-type Cx46 and Cx50 closed in 85 mseconds, a rate approximately 16% or 36% faster than the mixed channels containing Cx46 and S50P or homotypic Cx46 gap junctions, respectively. (G,H) Similarly, when a 120 mV V_j was applied, homotypic Cx46 channels closed in 126 mseconds (G), a mean channel closure time significantly slower then the 90 mseconds exhibited by the mixed channels containing wild-type Cx46 and S50P proteins (H) (P>0.05, ANOVA). Heteromeric channels containing both wild-type Cx46 and Cx50 (I), displayed a mean value of 75 mseconds, a closure speed that is significantly faster than either the homotypic Cx46 or heteromeric Cx46 and S50P channel (P>0.05, ANOVA). All means are the sum of four independent experiments.