Search

An inherited mutation in the KCNK18 gene has been shown to be associated with ‘migraine with aura’. This is a common, debilitating, recurrent headache disorder associated with transient and reversible focal neurological symptoms. KNCK18 encodes the TWIK-related spinal cord potassium channel (TRESK), a member of the K2P family of potassium channels. The F139WfsX24 mutation, segregates perfectly with typical migraine with aura in a large pedigree and functional characterization of this mutation demonstrates that it causes a complete loss of TRESK function and that the mutant subunit suppresses wild-type channel function through a dominant-negative effect, thus explaining the dominant penetrance of this allele (Lafreniere et al, doi:10.1038/nm.2216). This identifies a role for TRESK in the pathogenesis of typical migraine with aura and further supports the role of this channel as a potential therapeutic target. In this study we have examined the electrophysiological properties of other mutations identified in the human KCNK18 gene and find that several of these variants also produce a dramatic dominant-negative phenotype.

The TREK channel belongs to the superfamily of two-pore-domain potassium channels (K2P-channels) that are made up of four transmembrane segments (TM1 - TM4) and two pore-forming domains that are arranged in tandem. The activity of these channels is directly regulated by the intracellular pH, heat, polyunsaturated fatty acids, phospholipids and mechanical stretch. Currently little is known about the pore structure and how these different stimuli gate the pore in structural terms. To this end we employed systematic cysteine scanning mutagenesis on the four TM domains and functionally characterised these mutants in several respects: I) using chemical cysteine modification we identified pore lining residues, II) by measuring detailed pH dose response curve we identified residues involved in the pH gating mechanism and III) by studying different pore blocking compounds we identified potential blocker interacting residues. These sets of functional data will be evaluated in the context of structural models of the TREK channel pore in the closed and open state.

Specific stimuli (e.g. intracellular pH, PIP2) cause Kir channels to undergo a reversible transition between the closed and open state. The precise rearrangement of the pore structure (e.g. the TM regions) during these gating transitions is currently unknown in Kir channels owning to the lack of an open state crystal structure and validating functional data. We employed systematic alanine scanning mutagenesis on the entire Kir1.1 pore structure (residues 51 - 192) and determined the IC50 values for pH inhibition for each mutant. We reasoned that a shift in the IC50 value should represent a change in the free energy of the open state relative to the closed state assuming that the mutations did not affect the cytoplasmatic pH sensor directly. We identified 26 mutations that produced a marked shift in the pH sensitivity, intriguingly, 24 of these mutations increased the pH sensitivity suggesting that the open state is structurally more optimised than the closed state. Specifically, we expected that in the open state the IC50 shifting residues would interact with each other explaining why the alanine substitution would preferentially destabilise this state. We used this expectation as conceptual basis to develop scoring methods to evaluate structural models of Kir1.1 in the open state that we generated from existing crystallographic open state structures of other K+ channels (KvAP, NaK, KirBac3.1) using homology modelling and MD simulations. This analysis revealed an excellent agreement of our functional data with one particular open state model of Kir1.1. In this model more than 80% of the IC50 shifting residues are part of a tightly packed network of interacting residues that largely disintegrates upon channel closure.

Intracellular pH gates inwardly rectifying potassium (Kir) channels by controlling the reversible transition between the closed and open states. This gating mechanism underlies important aspects of Kir channel physiology and pathophysiology. H+ inhibition is thought to be triggered by protonation of residues within the cytoplasmic domains. This then causes major conformational rearrangement of the TM helices and slide helix resulting in closure of the gate at the helix bundle crossing. To identify residues important for this gating process we performed a systematic alanine scan in Kir1.1 channels over the entire transmembrane pore structure of the channel (residues 61 - 192) and measured pH sensitivity of the individual mutants in inside-out patches. We identified gating sensitive residues in both TM1 and TM2 that appear to make up an intrasubunit gating interface as well as a cluster of residues in the proximal part of the slide helix extending into TM1. Two highly conserved phenylalanines (F84, F88) in TM1 seem be of particular importance as they had dramatic effects on the pH gating kinetics. Assuming that the mutations do not affect the cytoplasmic pH sensor directly, a change in IC50 therefore represents a change in the stability of the closed state relative to the open state. Intriguingly, most of the gating sensitive mutations (17 out of 19) increased the IC50 for pH inhibition (from 6.4 (wild-type) up to 8.5) indicating that the mutations had a marked tendency to disturb the stability of the open state more severely than the closed state. This suggests that the open state in Kir channels is structurally more optimised than the closed state.

The molecular identity of ion channels which respond to PCO2/pH in the brain is still unclear. Heteromeric Kir4.1/Kir5.1 channels are highly sensitive to inhibition by intracellular pH and are widely expressed in several brainstem nuclei involved in cardiorespiratory control, including the locus coeruleus (LC). This has therefore led to a proposed role for these channels in neuronal CO2-chemosensitivity. In order to examine this we generated mutant mice lacking the Kir5.1 (Kcnj16) gene. We show that whilst LC neurons from Kcnj16(+/+) mice rapidly respond to cytoplasmic alkalinization and acidification, those from Kcnj16(−/−) mice display a dramatically reduced and delayed response. These results identify Kir5.1 as an important determinant of PCO2/pH sensitivity in locus coeruleus neurons and suggest that Kir5.1 is involved in the response to hypercapnic acidosis.