The Shaker K⁺ channel is a member of the Kv channel family, made up of four subunits each comprising six membrane embedded segments (S1-S6): S1-S4 form the voltage-sensing domain and S5-P-S6 the pore domain. During depolarisation S4, the principal component of the sensor, moves out of the membrane bilayer, which leads to opening of gates situated in the pore domain. Using the engineered metal and disulphide bridge approaches, we have shown that residue S357 in the S3-S4 linker (3 residues from S4) lies close to residue E418 at the top of S5 of an adjacent subunit in the closed state of the channel (Neale et al. 2003). Here we have investigated the interactions of residues downstream of S357, with E418.

Cysteine mutations were introduced at positions 358-362 in Shaker-IR. A second cysteine was introduced at 418. Single and double cysteine mutant channels were expressed in Xenopus oocytes and currents measured by two-electrode voltage clamp. The effects of 100 µM Cd²⁺ on the properties of these channels were examined. Effects on the double mutant channels that were significantly different from those on the single mutant channels were interpreted as Cd²⁺ binding between the engineered cysteines, reflecting their close proximity.

Cd²⁺ caused significant conductance loss (35-80% P < 0.05, Student’s unpaired t test) in 358C-418C, 360C-418C and 361C-418C double mutant channels, but not in the corresponding single mutant channels. These effects were seen only at depolarised potentials, indicating that depolarisation-induced conformational changes bring the engineered cysteines into close proximity. To investigate if Cd²⁺ forms a bridge between cysteines within the same subunit or from neighbouring subunits, we have made tandem dimer constructs, in which one cysteine was present in S4 of one protomer, and the second cysteine at 418 of the second protomer. The resulting constructs when expressed should form tetramers with the S4 and pore cysteines in diagonally located subunits. No effect of Cd²⁺ was seen on 360C-418C and 361C-418C tandem dimers, suggesting that the interactions require cysteines to be present in the same subunit. This is in contrast to the 357C-418C double mutant channel, where interactions between 357C and 418C, were found to be between the adjacent subunits.

Based on these data, we propose that the top of S4 moves from the top of S5 of a neighbouring subunit to its own subunit during gating, in a motion that involves a combination of vertical and horizontal translation. Such motions can be explained in terms of the recently solved structure of the archaebacterial Kv channel, KvAP (Jiang et al. 2003).

This work was funded by the Wellcome Trust.