Proceedings of The Physiological Society
University of Oxford (2011) Proc Physiol Soc 23, PC268
The solution structure of the EAG domain of hERG channels provides insight into the molecular basis for slow deactivation gating.
S. Thomson1, F. Muskett1, P. Stansfeld2, J. Mitcheson1
1. Cell Physiology and Pharmacology, University of Leicester, Leicester, United Kingdom. 2. University of Oxford, Oxford, United Kingdom.
The slow deactivation gating of hERG (Kv11.1) potassium channels is important for regulating the physiological time course and amplitude of the rapid delayed rectifier K+ current during the cardiac action potential. Inherited mutations that accelerate deactivation are linked to Long QT syndrome, a potentially lethal type of cardiac arrhythmia. hERG channels have large intracellular N- and C-termini. The EAG domain (residues 1-135) is known to be an important structural determinant of deactivation gating, but the precise mechanism is poorly understood. The C-terminus contains a structural domain with homology to a cyclic nucleotide binding domain (cNBD). In this study we solved the solution structure of the EAG domain using NMR spectroscopy, including the functionally critical first 26 residues (NT1-26). A previously undefined amphipathic helix from residues Gln11 to Gly24 was revealed and an extended highly dynamic region from residues 1-10. One face of NT1-26 was positively charged. Site directed mutagenesis was used to investigate if the positive face of NT1-26 and the α-helix were functionally important for deactivation gating. The mutants were expressed in Xenopus oocytes and currents measured using two electrode voltage clamp. The deactivation of hERG currents was fit with a single exponential function to determine the time constants (τ) for deactivation. All values given here are at a test potential of -70 mV and n = 5-7 oocytes. Alanine substitution of basic residues profoundly accelerated hERG current deactivation from mean τ values of 565±20 ms in wild-type hERG to 30±2 and 86±9 ms in R4A:R5A and R20A:K21A respectively. These deactivation time constants were similar to NTK hERG in which the whole N-terminus had been truncated (15±1 ms). In contrast, substitution of residues on the hydrophobic side of the amphipathic helix (F14A, L15A, and T17A) had minor effects on deactivation. Breaking or kinking the NT1-26 helix by substituting Pro at a central position (Ile18) resulted in fast deactivation with τ of 44±6 ms. Given the functional importance of basic residues on NT1-26 we searched for acidic regions on other structural domains of the channel. Homology models of the cNBD showed an extensive acidic patch with four-fold symmetry on the surface facing into the cytoplasm. Charge reversal mutations of acidic residues significantly (p<0.05) accelerated deactivation. The cNBD of hERG is proposed to be mechanically coupled to the activation gate. Based on our findings we propose a model for hERG deactivation gating in which the NT1-26 domain electrostatically interacts with the cNBD. This interaction stabilises the intracellular complex and is transduced to the pore to maintain the channel in the open conformation.
Where applicable, experiments conform with Society ethical requirements