Force open mutant channel properties in vitro. (A) Backbone trace of Kir2 channel with the side chain of the two residues (I177 and M181) forming the HBC in sticks and the mutated residue G178 in balls. PIP₂ binding site residues are shown in sticks, and green and bulk anionic lipid binding site residues are in sticks and cyan. (B) Sequence alignment of selected regions of K⁺ channels. The residues at the PIP₂ binding site are shown in green, and the residues important for secondary anionic lipids in cyan. The residues mutated in this study are highlighted in yellow, and the residues forming the HBC are highlighted in orange. The negatively charged residue in the inner cavity important for rectification, the so-called rectification controller, is shown in pink, and the conserved glycine hinge residues are highlighted in gray. (C) Mutant and control protein activity was assessed in the absence and presence of 0.1% PIP₂ in liposomes made with 10% POPG and 90% POPE. Data shows ⁸⁶Rb⁺ uptake normalized to the maximum uptake in valinomycin. ***, P < 0.001. (D) Phospholipid specificity was tested for KW and KW/GD proteins. Phosphatidylinositol 4,5-bisphosphate or phosphatidylinositol 4-phosphate was increased from zero to 1% in liposomes formed with 10% POPG/90% POPE. Data show ⁸⁶Rb⁺ uptake normalized to the maximum uptake in valinomycin. (E) Ion selectivity was assessed with different ions inside the proteoliposome, at a fixed lipid composition (0.1% PIP₂, 10% POPG, and 90% POPE). Data show ⁸⁶Rb⁺ uptake normalized to the uptake in K⁺ gradient. (F) Sensitivity to the pore-blocker spermine was assessed at a fixed lipid composition (0.1% PIP₂, 10% POPG, and 90% POPE). Data show ⁸⁶Rb⁺ uptake in the absence or presence of 10 µM spermine, normalized to the uptake in the absence of spermine. In C–F, error bars indicate mean ± SEM for n ≥ 3 independent protein preparations with one to three replicates for each preparation (see Materials and methods).