We expressed murine TASK-1 in oocytes from Xenopus anaesthetised by immersion in 0.3 % w/v MS222 and killed by destruction of the brain and spinal cord. We used two-electrode voltage clamp to investigate pH sensitivity and ionic selectivity. We mutated two residues, H98 in the first pore domain P1 and D204 in P2. We used a tandem dimer construct to mutate residues in one only of two subunits forming the ion channel. We fitted the pH sensitivity of wild-type TASK-1 assuming that only one His residue (H98) need be protonated for channels to be non-conducting (Ashmole et al. 2001).

The experimental results may be fitted using an equilibrium constant, K_a, for the protonation reaction of 9.55 X 10^-7 M (pK_a = 6.02) at +40 mV (70 mM [K⁺]_o). The effect was weakly voltage dependent. The requirement for only a single His to be protonated is supported by the finding that pH dependence is conserved in tandem dimers with a single H98D mutation. However pK_a was raised to 7.04 at +40 mV. This change is presumably caused by the introduction of a further acidic residue close to H98. The mutation H98D itself (i.e. in both subunits) reduces, but does not abolish, pH dependence, indicating that residues other than H98 contribute. Surprisingly, the mutation D204H forms channels whose activity is virtually independent of pH over the physiological range. While protonation of H98 appears to be necessary for channel closure, it may not be sufficient. Like Lopes et al. (2001), we believe that the response to acidification is not simply through channel blockage, but involves a gating process.

Mutation of H98 and of D204 also alters channel selectivity measured from the shift in reversal potential when 70 mM K⁺ is replaced by Rb⁺ or Na⁺. P_Rb/P_K was 0.80 ± 0.03 (n = 14; mean ± S.E.M.) in wild-type, 0.84 ± 0.04 (n = 19) in H98D and 1.06 ± 0.04 (n = 19) in D204H. At negative voltages, Rb⁺ currents were smaller than K⁺ currents in wild-type, but were larger in D204H. Channels became substantially Na⁺ permeable after mutation. In wild-type, P_Na/P_K was less than 0.09. But it was 0.39 ± 0.03 (n = 14) in H98D and 0.64 ± 0.04 (n = 18) in D204H. In D204H, Na⁺ currents were approximately 30 % the amplitude of K⁺ currents at negative voltages; at positive voltages, Na⁺ in the external solution blocked the efflux of K⁺. We find that twofold symmetry best supports both pH sensitivity over the physiological range and K⁺ selectivity.

We thank the BBSRC for support