Similarity of the ion occupations of K⁺ and Tl⁺ in the selectivity filter of a K⁺ channel, as verified by single-channel currents.

Figure 1.. Single-channel characteristics of KcvNTS with K+ and Tl+ as the permeant ions. (A,B) Current traces recorded at different membrane voltages in symmetrical (A) 100 mM KCl and (B) 100 mM TlNO3. Dashed lines indicate the baseline. (C,D) Apparent IV curves (Iapp) and open probabilities (PO) are shown as mean ± standard deviation. Error bars in (C) are too small to be visible. The number of independent experiments is given in brackets in the legend of panel (C).

Figure 2.. Representative amplitude histograms of single-channel recordings at −160 mV (A) and +160 mV (B) in 100 mM KCl (black) and 100 mM TlNO3 (orange). The approximate contributions of the three closed states in the Markov model are indicated. Smooth lines represent the fitted curves. (C) A Markov model is used for the fitting. The states and rate constants kXY are defined in the reaction scheme. (D) The closed state M is likely composed of two states M1 and M2 with similar dwell-time so they cannot be distinguished in the analysis. M1 and M2 have to be lumped together into state M for the extended beta distribution fit (see (C)). kOM,1 and kOM,2 can be separated by with the help of Eq. 1.

Figure 3.. Kinetics of KcvNTS wild type in KCl (black) and TlNO3 (orange) symmetrical solutions of (A,B) 100 mM and (D,E) 250 mM. (A,D) Open channel current Itrue. (B,E) Voltage dependence of O-M gating (filter gating). Filled symbols indicate the rate constant of channel closing, kOM, and open symbols the rate constant of channel opening, kMO. Crosses in (A) and (B) indicate control experiments in 100 mM KNO3, titrated with NMDG. (C) Permeation model used in our previous work [17], derived from Roux [2]. (F) Occupation probabilities of the filter binding sites in this model for KcvNTS wild type in symmetrical 100 mM KCl, data from Rauh et al. [17]. In (A,B,D,E), data are shown as mean ± standard deviation. Some error bars are too small to be visible. The number of independent experiments is given in brackets in the legends of panels (A,D).

Figure 4.. Combined effect of the mutation S42T and Tl+ on O-M gating. (A-C) Current traces recorded from the S42T mutant in symmetrical 100 mM KCl (A) and 100 mM TNO3 (B), and for comparison, from the wild type in 100 mM KCl (C), same traces as in Figure 1A. (D,E) Apparent IV curves (Iapp) and open probabilities (PO) are shown as mean ± standard deviation. The number of independent experiments is given in brackets in the legend of panel (E).

Figure 5.. (A,D) Comparison of Itrue for the wild type (circles) and the S42T mutant (triangles) in KCl (black) and TlNO3 (orange) at 100 and 250 mM. The kOM and kMO curves measured in the S42T mutant in (B,C) 100 mM and (E,F) 250 mM TlNO3 (orange triangles) coincide with those measured in KCl of the same concentrations in the wild type (black circles). Data are shown as mean ± standard deviation. Some error bars are too small to be visible. The number of independent experiments is given in brackets in the legends of panels (A,D).

Figure 6.. The ion on the external side determines kOM at negative voltages in asymmetrical K+/Tl+ solutions (A,D). the influence of the ions on kMO (B,E) is more complex. With Tl+ on the external side, kMO increases at negative voltages, even in comparison with symmetric Tl+, whereas cytosolic Tl+ causes a small decrease. (C,F) At negative voltages, Itrue is determined by the external ion (for the S42T mutant, no difference is observed between K+ and Tl+). At positive voltages, the situation is more complex. Data are shown as mean ± standard deviation. Some error bars are too small to be visible. The number of independent experiments is given in brackets in the legends of panels (A,D).