TRESK Background K+ Channel Is Inhibited by PAR-1/MARK Microtubule Affinity-Regulating Kinases in Xenopus Oocytes

PLOS ONE, Dec 2011

TRESK (TWIK-related spinal cord K+ channel, KCNK18) is a major background K+ channel of sensory neurons. Dominant-negative mutation of TRESK is linked to familial migraine. This important two-pore domain K+ channel is uniquely activated by calcineurin. The calcium/calmodulin-dependent protein phosphatase directly binds to the channel and activates TRESK current several-fold in Xenopus oocytes and HEK293 cells. We have recently shown that the kinase, which is responsible for the basal inhibition of the K+ current, is sensitive to the adaptor protein 14-3-3. Therefore we have examined the effect of the 14-3-3-inhibited PAR-1/MARK, microtubule-associated-protein/microtubule affinity-regulating kinase on TRESK in the Xenopus oocyte expression system. MARK1, MARK2 and MARK3 accelerated the return of TRESK current to the resting state after the calcium-dependent activation. Several other serine-threonine kinase types, generally involved in the modulation of other ion channels, failed to influence TRESK current recovery. MARK2 phosphorylated the primary determinant of regulation, the cluster of three adjacent serine residues (S274, 276 and 279) in the intracellular loop of mouse TRESK. In contrast, serine 264, the 14-3-3-binding site of TRESK, was not phosphorylated by the kinase. Thus MARK2 selectively inhibits TRESK activity via the S274/276/279 cluster, but does not affect the direct recruitment of 14-3-3 to the channel. TRESK is the first example of an ion channel phosphorylated by the dynamically membrane-localized MARK kinases, also known as general determinants of cellular polarity. These results raise the possibility that microtubule dynamics is coupled to the regulation of excitability in the neurons, which express TRESK background potassium channel.

TRESK Background K+ Channel Is Inhibited by PAR-1/MARK Microtubule Affinity-Regulating Kinases in Xenopus Oocytes

Czirjak G (2011) TRESK Background K+ Channel Is Inhibited by PAR-1/MARK Microtubule Affinity-Regulating Kinases in Xenopus Oocytes. PLoS ONE 6(12): e28119. doi:10.1371/journal.pone.0028119 + TRESK Background K Channel Is Inhibited by PAR-1/ MARK Microtubule Affinity-Regulating Kinases in Xenopus Oocytes Gabriella Braun 0 Bala zs Nemcsics 0 Pe ter Enyedi 0 Ga bor Czirja k 0 Bernard Attali, Sackler Medical School, Tel Aviv University, Israel 0 Department of Physiology, Semmelweis University , Budapest , Hungary TRESK (TWIK-related spinal cord K+ channel, KCNK18) is a major background K+ channel of sensory neurons. Dominantnegative mutation of TRESK is linked to familial migraine. This important two-pore domain K+ channel is uniquely activated by calcineurin. The calcium/calmodulin-dependent protein phosphatase directly binds to the channel and activates TRESK current several-fold in Xenopus oocytes and HEK293 cells. We have recently shown that the kinase, which is responsible for the basal inhibition of the K+ current, is sensitive to the adaptor protein 14-3-3. Therefore we have examined the effect of the 14-3-3-inhibited PAR-1/MARK, microtubule-associated-protein/microtubule affinity-regulating kinase on TRESK in the Xenopus oocyte expression system. MARK1, MARK2 and MARK3 accelerated the return of TRESK current to the resting state after the calcium-dependent activation. Several other serine-threonine kinase types, generally involved in the modulation of other ion channels, failed to influence TRESK current recovery. MARK2 phosphorylated the primary determinant of regulation, the cluster of three adjacent serine residues (S274, 276 and 279) in the intracellular loop of mouse TRESK. In contrast, serine 264, the 14-3-3-binding site of TRESK, was not phosphorylated by the kinase. Thus MARK2 selectively inhibits TRESK activity via the S274/276/279 cluster, but does not affect the direct recruitment of 14-3-3 to the channel. TRESK is the first example of an ion channel phosphorylated by the dynamically membranelocalized MARK kinases, also known as general determinants of cellular polarity. These results raise the possibility that microtubule dynamics is coupled to the regulation of excitability in the neurons, which express TRESK background potassium channel. - TRESK is abundantly expressed in dorsal root ganglion (DRG) neurons and has been suggested to play an important role in pain disorders [15]. TRESK is the target of sanshool, the paresthetic and counter-irritant ingredient of the traditional Chinese medicine, Sichuan pepper [6,7]. The channel has recently attracted particular attention, because its dominant-negative mutation was reported to be linked to familial migraine with aura [8]. These findings indicate the importance of TRESK in pain control and points to the need for better understanding of the regulatory properties of the channel. TRESK regulation is distinguished within the K2P channel family by the unique sensitivity to the cytoplasmic calcium signal. The calcium/calmodulin-dependent protein phosphatase calcineurin activates TRESK 515-fold in Xenopus oocytes [9]. Stimulation of Gq protein-coupled receptors activated TRESK by 4080% in COS-7 cells under whole-cell patch clamp conditions [10,11]. Whole-cell TRESK current in native cells has not been reliably measured, although several studies examined TRESK in isolated DRG neurons [5,8,1013]. In the absence of specific inhibitors, separation of native whole-cell TRESK current from the other endogenous background K+ currents remains a challenge to be solved in the future. When cell-attached patches containing TRESK channels were painstakingly selected from DRG neurons, single channel activity increased by 3080% in response to receptor stimulation [11]. The mechanism of TRESK activation in mammalian cells, and the cause of the apparently smaller stimulation of the current in the mammalian cell lines than in the Xenopus system have not yet been investigated. We have recently realized that two inhibitory kinase pathways converge on TRESK [14]. The two pathways have different target residues in the intracellular loop of the channel. Protein kinase A phosphorylates the second serine in the conserved RSNSCPE sequence (S264 in mouse and S252 in human TRESK), thereby recruits the adaptor protein 14-3-3 to this motif [15], and exerts auxiliary channel inhibition [14]. However, the major inhibitory pathway targets the S274/276/279 cluster; RLSCSILSNLD in the mouse, corresponding to RLSYSIISNLD (S262/264/267) in human TRESK. Intriguingly, this pathway was shown to be inhibited by 14-3-3 even if the direct binding of the adapter protein to TRESK was abrogated [14]. The major aim of our present study was to identify the kinase, which phosphorylates the S274/276/279 cluster and accordingly inhibits TRESK, when expressed in the Xenopus oocyte system. Materials and Methods Plasmids and reagents The cloning of human and mouse TRESK cDNAs [9] and S264E mutant mouse TRESK [14] were previously described. Mouse TRESK was subcloned to pIRES-CD8 vector [16] for transfection of HEK293 cells. Human embryonic kidney (HEK293) cell line (ATCC-CRL-1573) was purchased from LGC Standards GmbH (Wesel, Germany). The AMPK-related kinase and tau cDNAs were amplified with RT-PCR. Total RNAs were purified with TRIzol reagent (Invitrogen, Carlsbad, CA). Reverse transcription was performed with MMLV-RT (Revertaid, Fermentas, Vilnius, Lithuania) from mouse brain (BRSK1, MARK1, MARK2, MARK3, MARK4, NUAK1, tau), embryo body (SIK1(1343)), testis (AMPKa1) or placenta (MELK) total RNAs. MARK1 and MELK PCR products were amplified with Ultra Pfu (Stratagen, La Jolla, CA), while those of the other kinases with Pfu polymerase (Fermentas). We have cloned isoform 2 of MARK2 (722 amino acids, Genebank NP_001073857), and used this protein throughout the study. For primer sequences, cloning sites and PCR protocols see Figure S1. All kinase cDNAs were cloned to pXEN vector (Genebank EU267939), and verified by automatic sequencing. Different mutant versions of the kinases were produced with QuikChange site directed mutagenesis (Stratagen). For primer sequences see Figure S1. MARK2 T208E or T208E/T539A mutants were also subcloned into pGEX2TK4T1 and pET32DKpn [15] vectors for the production of GST- or Trx-His6-tagged versions of the kinase in E. coli. (In pGEX2TK4T1 the EcoRI-PstI fragment of pGEX-4T-1 was cloned into pGEX-2TK, Amersham Biosciences, Little Chalfont, UK.) Tau coding sequence was cloned into pGEX-4T-1. Cloning and purification of GSTTRESKloop, GST-TRESKloop-TAPtag and different versions of TRESKloop-His8 protein were previously described [15,17]. These proteins are hydrophobic; they are insoluble under nondenaturing conditions. Therefore, after prurification from bacterial lysates they were kept immobilized on the affinity matrix (on glutathione or Ni-NTA resins) and they were added in this form to the kinase reaction. Ionomycin (calcium sal (...truncated)


This is a preview of a remote PDF: http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0028119&type=printable
Article home page: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0028119

Gabriella Braun, Balázs Nemcsics, Péter Enyedi, Gábor Czirják. TRESK Background K+ Channel Is Inhibited by PAR-1/MARK Microtubule Affinity-Regulating Kinases in Xenopus Oocytes, PLOS ONE, 2011, 12, DOI: 10.1371/journal.pone.0028119