Pore Helix-S6 Interactions are Critical in Governing KCNQ3 Amplitudes

Abstract

Two mechanisms have been postulated as underlying the small KCNQ3 homomeric currents, compared with other KCNQ homomers, or KCNQ2/3 heteromers. The first involves differential channel expression, governed by the assembly domain of the distal C-terminus (Schwake et al., 2006. J. Neuroscience), centered on the “D helix,” whose KCNQ4 crystal structure revealed a “coiled-coil” suggested to be much more favorable for tetramerization of KCNQ2/3 heteromers and KCNQ4 homomers than for KCNQ3 homomers (Howard et al., 2007. Neuron). The second suggests similar expression of KCNQ channels, but a KCNQ3 pore that is particularly unstable, leading to most KCNQ3 homomers being dormant, whereas mutation of an intracellular pore-helix residue, A315, to a hydrophilic T/S boosted currents by >25-fold (Zaika et al., 2008. Biophys J; Exteberria et al., 2004. J. Neuroscience). Pore instability is thought to underlie “C-type” inactivation of K+ channels, including KcsA, for which disruption of the interaction between F103 (S6) and the pore helix (T74, T75) stabilized the pore (Cuello et al., 2010. Nature). We found mutations at the analogous position in KCNQ3 (F344) dramatically decreased KCNQ3 currents, and TIRF showed negligible effects on membrane expression. Homology modeling of wild-type and mutant KCNQ3 suggest the decrease of the current in F344 mutants is due to a disruption of the interaction between F344 (S6) and the A315 residue. Finally, native PAGE revealed H2O2-enhanced oligomerization of KCNQ4 subunits at C643 at the end of the D-helix. However, H2O2-mediated enhancement of KCNQ4, previously showed to localize to a cysteine triplet in the S2-S3 linker (Gamper et al., 2006. EMBO J), was identical in the C643A mutant. Our results suggest a secondary role of the C-terminus, compared to pore-helix-S6 interactions, governing KCNQ3 amplitudes, and that variable surface expression among KCNQ channels plays only a minor role.