An extracellular drug binding site of potassium channels THIK-1 and THIK-2
Background: THIK-1, THIK-2 and TREK-1 and all belong to family of two-pore-domain potassium channel (K2P channels). THIK-2 was until recently regarded as ‘silent’ potassium channels. 3-isobutyl-1-methylxanthine (IBMX) is normally used as inhibitor of phosphodiesterase, resulting in an increase of cA...
|Online Access:||PDF Full Text|
No Tags, Be the first to tag this record!
|Summary:||Background: THIK-1, THIK-2 and TREK-1 and all belong to family of two-pore-domain potassium channel (K2P channels). THIK-2 was until recently regarded as ‘silent’ potassium channels. 3-isobutyl-1-methylxanthine (IBMX) is normally used as inhibitor of phosphodiesterase, resulting in an increase of cAMP levels in the cytosol.
Aims: Identification of an extracellular drug binding site on THIK-1 and THIK-2.
Methods: Whole cell recording patch clamp measurements in mammalian cells was used to analyze K2P channels mentioned above. Chemicals such as IBMX, forskolin and cAMP were used intracellularly (via the pipette solution) and/or extracellularly (via the bath solution). To identify the binding site of IBMX on THIK-1 we mutated all amino acids of the helical cap one by one and screened for changes in IBMX sensitivity of the channels. To analyze the surface expression of the channel we used HA-tagged THIK-2 and thus quantified the copy number of the channels at the cell membrane using an antibody-based assay.
Results: We found that IBMX can rapidly inhibit both the inward and outward currents carried by THIK-1 channels; the IC50 of this effect was about 120 μM. The application of H89 (PKA inhibitor) and forskolin (PKA activator) did not modify the effects of IBMX on the channel. Application of 100 μM intracellular cAMP almost completely inhibited TREK-1 current but not THIK-1 current, indicating that the effect of IBMX in THIK-1 is not mediated by cAMP. Finally, we found that IBMX blocks THIK-1 currents only if it is applied extracellularly. By mutating all of the helical cap amino acids, we found that the arginine to alanine mutant of THIK-1 (THIK-1R92A) had a lower sensitivity to IBMX. Mutation of the arginine at position 92 to glutamate or glutamine reduced the sensitivity to IBMX even further. R92 is localized to the linker region between cap helix 2 (C2) and the pore helix (P1). Part of the linker region is not visible in the crystal structures. R92 is at the end of the unstructured region.Compared to THIK-1, the 'silent' channel THIK-2 has an additional domain at its N-terminus (residues 6-24) which contains a putative retention signal (RRR). Removal of this additional domain (mutant THIK-2Δ6-24) or mutation of the RRR motif to AAA (THIK-2AAA mutant) gave rise to a measurable potassium current. Furthermore, the surface expression of the reporter protein CD74 containing the AAA mutated N-terminus of THIK-2 was more than threefold larger than the analogous reporter protein containing the wild type N-terminus of THIK-2 (RRR). These data indicate that the ER retention/retrieval signal RRR can prevent the THIK-2 export to the cell membrane, leading to the silence of the channel. In addition, we found that THIK-2 currents can also be blocked by application of IBMX from the extracellular side.
Conclusions: IMBX can block TREK-1 channels though the PKA pathway, it also can bind to the extracellular side of THIK-1 or THIK-2, leading to a direct block of the channels. This describes a novel effect of IBXM on K2P channels. The IC50 of the direct effect of IBMX on THIK-1 channels was about 120 μM. Our results suggest that arginine 92 of THIK-1 and the C2-P1 linker region of K2P channels play an important role in the binding of IBMX, and perhaps other more potent drugs, to the channel.|