Fig. 5: Structure analysis of BsKtrAB.

a Surface representation (gray) of BsKtrB pore in the ATP- (Structure II, left panel) and ADP-bound (Structure III, right panel) BsKtrAB complexes, calculated by HOLLOW. D1-D4 domains and D1M2b helix are highlighted with the same color code as in Fig. 1a. The D2 ___domain of BsKtrB is transparent for clarity. The side chains of Phe91 and Arg417 are shown in stick representation. b Pore radius as a function of the distance along the axis of BsKtrB pore, calculated by HOLE, is plotted with respect to the positions of selectivity filter and intramembrane loop shown in a. c, d A close-up view of the hydrogen bond interactions in the vicinity of the intramembrane loop and D1M2b helix of BsKtrB at (c) ATP- and (d) ADP-bound states as magnified from the dashed boxes defined in a. The side chains of the residues involved in the hydrogen bond interactions (gray dashed lines) are shown in stick representation. e The hydrogen bonding probabilities of Arg417-Gly87 and Arg417-Thr310 in the presence of ATP (black) and ADP (gray) are plotted on the basis of molecular dynamics simulation. Data represent the mean ± s.d. with n = 7 (ATP) and n = 12 (ADP) independent experimental replicates. Statistical analyses were performed using two-way ANOVA. f Left panel, fluorescence-based K⁺ flux assays using wild-type BsKtrAB and KtrAB_F91A in the presence of ATP or ADP. The addition of H⁺ ionophore CCCP and valinomycin are indicated as black and white arrows. Right panel, plot of flux rate constants, which are calculated by fitting the data (100–500 s) to a one-phase decay model. Data represent the mean ± s.d. with n = 5 independent experimental replicates. Statistical analyses were performed by one-way ANOVA, and n.s. indicates no significance (p ≥ 0.05). g The force required for K⁺ to permeate cross the pore as a function of the distance along the pore analyzed using SMD simulation. Source data for (b, e, g, f) are provided as a Source Data file.