Quantum Calculations Show Caution Is Needed In Interpreting Methanethiosulfonate Accessibility Experiments On Ion Channels

Abstract

Standard models of ion channel voltage gating require substantial movement of one transmembrane segment, S4, of the voltage sensing domain. Evidence comes from the accessibility to external methanethiosulfonate (MTS) reagents of the positively charged arginine residues (R) on this segment. These are first mutated to cysteines (C), which in turn react with MTS reagents; it is assumed that the C is passively carried in the S4 movement, becoming accessible on one side or the other of the membrane. However, the Rs were salt bridged to negatively charged residues on other transmembrane segments, or hydrogen bonded, while C reacts as a negative ion. The space available for MTS is fairly close to the difference in volume between the large R residue and much smaller C, so the MTS is not severely sterically hindered. A reagent molecule can reach a C side chain; the C can react if not repelled by a negative charge from the amino acid to which the R had been salt bridged. Nearby protons may also make reaction possible unless the C itself is protonated. Therefore interpretation of the C substitution results requires reconsideration. To test the idea we have done quantum calculations on part of a mutated S4 and the nearby section of the channel. The mutation is R300C of the 2A79/3Lut structure, a mutation that would be done to test MTS reagent access; there is a large cavity where the R is replaced by C. Two quantum calculations show a substantial difference in the structure of this cavity with 2 water molecules compared to 4. This suggests that the structure, and presumably reaction probability, could depend on water molecules, very likely also protons, in or near the cavity that the R300C mutation produces.