Location and Orientation of minK within the IKsPotassium Channel Complex

The slowly activating cardiac potassium current (IKs) is generated by a heteromultimeric potassium channel complex consisting of pore-forming (KvLQT1) and accessory (minK) subunits belonging to the KCNQ and KCNEgene families, respectively. Evidence indicating that minK residues line the IKs pore originates from the observation that two minK cysteine mutants (G55C and F54C) render IKsCd2+-sensitive. We have identified a single cysteine residue in the KvLQT1 S6 segment (Cys-331) that contributes to Cd2+ coordination in conjunction with cysteine residues engineered into the minK transmembrane domain. This observation indicates that minK resides in close proximity to S6 in the IKs channel complex. On the basis of homology modeling that compares the KvLQT1 S6 segment with the structure of the bacterial potassium channel KcsA, we predict that the sulfhydryl side chain of Cys-331 projects away from the central axis of the KvLQT1 pore and suggest that minK resides outside of the permeation pathway. A preliminary model illustrating the orientation of minK with S6 was validated by successful prediction of a novel Cd2+ binding site created within the IKs channel complex by engineering additional cysteine residues into both subunits. Our results indicate the location and orientation of minK within the IKs channel complex and further suggest that Cd2+ exerts its effect on IKs through an allosteric mechanism rather than direct pore blockade. The slowly activating cardiac potassium current (IKs) is generated by a heteromultimeric potassium channel complex consisting of pore-forming (KvLQT1) and accessory (minK) subunits belonging to the KCNQ and KCNEgene families, respectively. Evidence indicating that minK residues line the IKs pore originates from the observation that two minK cysteine mutants (G55C and F54C) render IKsCd2+-sensitive. We have identified a single cysteine residue in the KvLQT1 S6 segment (Cys-331) that contributes to Cd2+ coordination in conjunction with cysteine residues engineered into the minK transmembrane domain. This observation indicates that minK resides in close proximity to S6 in the IKs channel complex. On the basis of homology modeling that compares the KvLQT1 S6 segment with the structure of the bacterial potassium channel KcsA, we predict that the sulfhydryl side chain of Cys-331 projects away from the central axis of the KvLQT1 pore and suggest that minK resides outside of the permeation pathway. A preliminary model illustrating the orientation of minK with S6 was validated by successful prediction of a novel Cd2+ binding site created within the IKs channel complex by engineering additional cysteine residues into both subunits. Our results indicate the location and orientation of minK within the IKs channel complex and further suggest that Cd2+ exerts its effect on IKs through an allosteric mechanism rather than direct pore blockade. slowly activating cardiac potassium current Xenopus KvLQT1 Ion channels are protein macromolecules that serve a vast array of physiological functions. Many types, including voltage-gated potassium channels, are assembled as heteromultimeric complexes including distinct pore-forming and accessory or modulatory subunits. The manner in which these complexes are formed and the structural basis by which subunits interact are under-studied areas of ion channel biology. Members of a new class of voltage-gated potassium channels encoded by the KCNQ gene subfamily have been implicated in inherited forms of cardiac arrhythmia (1Wang Q. Curran M. E. Splawski I. Burn T. C. Millholland J. M. VanRaay T. J. Shen J. Timothy K. W. Vincent G. M. de Jager T. Schwartz P. J. Toubin J. A. Moss A. J. Atkinson D. L. Landes G. M. Connors T. D. Keating M. T. Nat. Genet. 1996; 12: 17-23Crossref PubMed Scopus (1496) Google Scholar, 2Neyroud N. Tesson F. Denjoy I. Leibovici M. Donger C. Barhanin J. Fauré S. Gary F. Coumel P. Petit C. Schwartz K. Guicheney P. Nat. Genet. 1997; 15: 186-189Crossref PubMed Scopus (747) Google Scholar), epilepsy (3Charlier C. Singh N. A. Ryan S. G. Lewis T. B. Reus B. E. Leach R. J. Leppert M. Nat. Genet. 1998; 18: 53-55Crossref PubMed Scopus (817) Google Scholar, 4Singh N. A. Charlier C. Stauffer D. DuPont B. R. Leach R. J. Melis R. Ronen G. M. Bjerre I. Quattlebaum T. Murphy J. V. McHarg M. L. Gagnon D. Rosales T. O. Peiffer A. Anderson V. E. Leppert M. Nat. Genet. 1998; 18: 25-29Crossref PubMed Scopus (1026) Google Scholar), and deafness (5Kubisch C. Schroeder B. C. Friedrich T. Lutjohann B. El-Amraoui A. Marlin S. Petit C. Jentsch T. J. Cell. 1999; 96: 437-446Abstract Full Text Full Text PDF PubMed Scopus (679) Google Scholar, 6Jentsch T. J. Nat. Rev. Neurosci. 2000; 1: 21-30Crossref PubMed Scopus (687) Google Scholar). KvLQT1 (KCNQ1), the first identified member of this subfamily, is expressed predominantly in heart, where it coassembles with minK (7Barhanin J. Lesage F. Guillemare E. Fink M. Lazdunski M. Romey G. Nature. 1996; 384: 78-80Crossref PubMed Scopus (1392) Google Scholar, 8Sanguinetti M. C. Curran M. E. Zou A. Shen J. Spector P. S. Atkinson D. L. Keating M. T. Nature. 1996; 384: 80-83Crossref PubMed Scopus (1515) Google Scholar), a 129-amino acid residue accessory subunit to form a channel complex that generates the slowly activating cardiac potassium current (IKs), 1an important determinant of myocardial repolarization. Mutations in KvLQT1 are the most frequent cause of congenital long QT syndrome (9Splawski I. Shen J. Timothy K. W. Lehmann M. H. Priori S. Robinson J. L. Moss A. J. Schwartz P. J. Towbin J. A. Vincent G. M. Keating M. T. Circulation. 2000; 102: 1178-1185Crossref PubMed Scopus (1083) Google Scholar, 10Schwartz P. J. Priori S. G. Spazzolini C. Moss A. J. Vincent G. M. Napolitano C. Denjoy I. Guicheney P. Breithardt G. Keating M. T. Towbin J. A. Beggs A. H. Brink P. Wilde A. A. Toivonen L. Zareba W. Robinson J. L. Timothy K. W. Corfield V. Wattanasirichaigoon D. Corbett C. Haverkamp W. Schulze-Bahr E. Lehmann M. H. Schwartz K. Coumel P. Bloise R. Circulation. 2001; 103: 89-95Crossref PubMed Scopus (1465) Google Scholar), an inherited disorder characterized by syncope, cardiac arrhythmias, and sudden death (11Schwartz P. J. Curr. Probl. Cardiol. 1997; 22: 297-351Crossref PubMed Google Scholar, 12Vincent G. M. Annu. Rev. Med. 1998; 49: 263-274Crossref PubMed Scopus (160) Google Scholar). Furthermore, minK belongs to the KCNE family of potassium channel subunits that includes the minK-related proteins that have also been implicated in cardiac arrhythmias (13Schulze-Bahr E. Wang Q. Wedekind H. Haverkamp W. Chen Q. Sun Y. Rubie C. Hordt M. Towbin J. A. Borggrefe M. Assmann G. Qu X. Somberg J. C. Breithardt G. Oberti C. Funke H. Nat. Genet. 1997; 17: 267-268Crossref PubMed Scopus (369) Google Scholar, 14Abbott G. W. Sesti F. Splawski I. Buck M. E. Lehmann M. H. Timothy K. W. Keating M. T. Goldstein S. A. N. Cell. 1999; 97: 175-187Abstract Full Text Full Text PDF PubMed Scopus (1173) Google Scholar, 15Sesti F. Abbott G. W. Wei J. Murray K. T. Saksena S. Schwartz P. J. Priori S. G. Roden D. M. George Jr. , A. L. Goldstein S. A. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10613-10618Crossref PubMed Scopus (442) Google Scholar) and inherited forms of periodic paralysis (16Abbott G. W. Butler M. H. Bendahhou S. Dalakas M. C. Ptacek L. J. Goldstein S. A. N. Cell. 2001; 104: 217-231Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar). Recombinant KCNQ/KCNE potassium channels are excellent tools with which to study physiologically important subunit-subunit interactions. Previous work by many investigators has provided provocative ideas regarding how minK interacts with KvLQT1 and which regions of KvLQT1 are involved in minK-mediated modulation. Romeyet al. (17Romey G. Attali B. Chouabe C. Abitbol I. Guillemare E. Barhanin J. Lazdunski M. J. Biol. Chem. 1997; 272: 16713-16716Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar) used yeast two-hybrid and immunodetection analysis to demonstrate that the KvLQT1 pore loop may interact directly with the minK carboxyl terminus, although surprisingly, no other intersubunit interactions were detected. Tai and Goldstein (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar) used scanning cysteine mutagenesis to demonstrate that cysteine substitution of amino acids within the minK transmembrane domain rendered IKs susceptible to partial block by external Cd2+. In this study, Cd2+ inhibition of the potassium current appeared to occur by direct pore block, suggesting that the minK transmembrane domain may line the permeation pathway of the channel complex. Recently, Kurokawa et al. (22Kurokawa J. Motoike H. K. Kass R. S. J. Gen. Physiol. 2001; 117: 43-52Crossref PubMed Scopus (42) Google Scholar) argued against the idea that minK resides within the IKspore region on the basis of their observation that channels remain Cd2+-sensitive during high affinity pore block by tetraethylammonium. We have recently shown that association of minK with KvLQT1 is mediated, in part, by residues within its transmembrane segment (19Tapper A. R. George Jr. , A. L. J. Gen. Physiol. 2000; 116: 379-390Crossref PubMed Scopus (79) Google Scholar). Despite these efforts, fundamental questions regarding the location and orientation of minK within the IKs channel complex remain unanswered. We have now determined that a cysteine residue native to the KvLQT1 S6 transmembrane segment (Cys-331) is necessary to form a Cd2+binding site in conjunction with engineered cysteines in the minK transmembrane domain. These data indicate that the minK transmembrane segment must reside in close proximity to the KvLQT1 S6 segment in the IKs channel complex. By homology with the structure of the bacterial potassium channel KcsA, the sulfhydryl side chain of Cys-331 projects away from the central axis of the KvLQT1 pore, and this argues that minK resides outside of the permeation pathway. Using a helical wheel model for the orientation of minK in relation to S6, we successfully predicted a novel Cd2+ binding site that could be created by engineering additional cysteine residues into both subunits. Our findings establish spatial relationships between two important families of potassium channel subunits and provide a new framework for interpreting data regarding structure and function inKCNQ/KCNE channels. Mutations in KvLQT1 and minK were introduced using site-directed polymerase chain reaction mutagenesis as previously described (19Tapper A. R. George Jr. , A. L. J. Gen. Physiol. 2000; 116: 379-390Crossref PubMed Scopus (79) Google Scholar). KvLQT1 and minK mutant constructs were unidirectionally ligated into plasmid vectors pSP64T and pRc/CMV, respectively. RNA from all KvLQT1 and minK contructs were transcribed in vitro from EcoRI- orXbaI-linearized DNA templates using Sp6 or T7 RNA polymerase and other reagents included in the mMessage mMachine transcription system (Ambion Corp. ). RNA size and integrity were evaluated by formaldehyde-agarose gel electrophoresis, and full-length RNA concentrations were estimated by comparison with a 0. 24–9. 5-kb RNA ladder (Life Technologies, Inc. ). Currents were recorded from oocytes at room temperature 2–5 days after injection using a two-microelectrode voltage-clamp technique with an OC-725B amplifier (Warner Instruments Corp. ). Pipettes were filled with 3 m KCl and had 0. 5–2mΩ resistances. Oocytes were bathed in ND-96 containing (in mm): 96 NaCl, 2 KCl, 1. 8 CaCl2, 1 MgCl2, and 5 HEPES, pH 7. 5. For Cd2+experiments, a modified ND-96 solution in which NaCl was replaced isotonically with CdCl2 was used. In these experiments, oocytes were depolarized to +20 mV for 10 s from a holding potential of −80 mV, and this was repeated at 30-s intervals. Once a steady-state current level was achieved, Cd2+ was perfused into the bath, and repetitive pulsing was continued to monitor the effect. Typically, Cd2+ elicited a maximum effect after 8–11 pulses (320–440 s). Washout was performed by bath perfusion with ND-96 after a plateau in Cd2+-induced current reduction was reached. To minimize contamination of endogenous XenopusKvLQT1 (xKvLQT1), only currents larger than 1 μA were selected for use in experiments. In addition, xKvLQT1 levels were monitored in each batch of oocytes by injecting minK cRNA into control cells. For each set of experiments, IKs derived from xKvLQT1 never exceeded 0. 2 μA when recorded at the end of a 10-s test depolarization to +20 mV. Data were recorded using the pCLAMP-6 software program (Axon Instruments, Inc. ), filtered at 200 Hz, and digitized at 1 kHz. Data were analyzed and plotted using a combination of pCLAMP and Origin (Microcal) software packages. Normalized isochronal voltage-activation relationships were obtained by measuring current 2 s after the onset of depolarizing test pulses between −50 and +60 mV from a holding potential of −80 mV. Data were fit with a Boltzmann function of the form: 1/1 + exp (V −V½app) /kapp, where V½app is the apparent half-maximal activation voltage and kapp is the apparent slope factor. Other data are presented as means ± S. E. , and comparisons were made using Student's t test with significance assumed at the p < 0. 05 A homology model of the KvLQT1 pore was by the KvLQT1 amino acid the pore loop through the S6 transmembrane segment with the obtained from the Data and N. M. C. 1997; 18: PubMed Scopus Google Scholar) were performed by the The model was rendered using the within Tai and Goldstein (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar) that cysteine substitution of two residues within the minK transmembrane segment and expressed IKs channels with with to partial block by external Cd2+. this effect of Cd2+ on channels expressed depolarization to +20 mV, oocytes KvLQT1 with minK slowly activating potassium currents that are to external Cd2+ A By oocytes KvLQT1 with of two minK cysteine mutants or have currents that be and by external Cd2+. a to currents derived from or minK and KvLQT1 channels were by an of ± and ± respectively. with only ± for channels. These data indicate that of KvLQT1 with or minK a Cd2+ coordination site that is or in channels. In complex protein binding of a occur when two or Lewis amino acid side as in are in close proximity J. M. of Scholar). In channel complexes containing or we that a cysteine residue native to KvLQT1 with engineered minK cysteines to Cd2+. To test we native KvLQT1 the mutants with or and determined the effect of Cd2+ on the channels. KvLQT1 endogenous and Cys-331 are within transmembrane and could reside to the domain of The cysteine residues are within regions of the protein and were to be for To of these residues are for the effect of we each residue to and and characterized the channels in the or of the minK cysteine expressed all KvLQT1 mutants formed channels that activating potassium currents in to depolarization mutant also a in currents that is of The voltage of activation determined for each mutant was also to that of although have an in activation with as as a of and of minK with each mutant in of of and a in the apparent voltage of activation and minK of and to a than KvLQT1 these results indicate that the function of the voltage at half-maximal activation and slope for mutant and KvLQT1 expressed with or determined in the and of ± ± ± ± ± ± ± ± + ± ± ± KvLQT1 + ± ± ± < KvLQT1 + ± ± ± < KvLQT1 + ± ± ± determined in the and of p < in a new of or with or minK currents that were by external of or with in currents The of steady-state current at +20 mV in these was ± and ± for expressed with and with ± ± and ± for KvLQT1 expressed with and at a Cd2+ the of block by channel complexes containing was than for channels containing KvLQT1 after of 2 external KvLQT1 and ± KvLQT1 and ± KvLQT1 and ± and ± and and ± data are from at results indicate that the effect of Cd2+ on IKs channel complexes containing or is on Cys-331 in KvLQT1 and that this residue with a cysteine in the minK transmembrane domain to form a ion coordination that the in with structure is L. J. J. 1998; PubMed Scopus Google Scholar), results indicate that minK must reside in close proximity to the KvLQT1 S6 segment in the IKs channel complex. are data both and the idea that minK may line the pore of IKs channels (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar, J. Motoike H. K. Kass R. S. J. Gen. Physiol. 2001; 117: 43-52Crossref PubMed Scopus (42) Google Scholar). In of data suggesting of the minK transmembrane domain with the S6 segment of the spatial orientation of the Cys-331 side chain to the central pore minK resides within or external to the pathway. Using the structure of the bacterial potassium channel J. M. A. J. M. R. 1998; PubMed Scopus Google Scholar), we created a KvLQT1 homology model to the of Cys-331 N. M. C. 1997; 18: PubMed Scopus Google Scholar). that the predicted of this side chain in model projects away from the central axis of the permeation pathway. This that to form a ion binding site between Cys-331 and the accessory subunit must into close with the region of the S6 transmembrane segment to the away from the channel that the minK transmembrane segment also forms an this further which helical must be directly to of a Cd2+ coordination site between minK or and the KvLQT1 S6 segment on helical wheel we predicted that a novel site could be engineered into the IKs channel complex by cysteine residues in the minK transmembrane domain and KvLQT1 S6 segment that were in the orientation as in minK and Cys-331 in To test we cysteines at minK and KvLQT1 in the of the of these residues are predicted to reside helical to minK amino acid and KvLQT1 residue respectively. a cysteine at of minK render the channel Cd2+-sensitive of minK in Xenopus oocytes generated IKs indicating that Cd2+ to the endogenous KvLQT1 in with previously (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar). of KvLQT1 with minK in the of currents derived from of KvLQT1 with minK were and by external reduction was ± In addition, the Cd2+ to 2 a ± current indicating that the Cd2+ effect is These data indicate that and are close to form a novel ion binding site and the predicted location and orientation of minK within the IKs channel complex. KvLQT1 with minK in the to form a channel complex for the slowly activating potassium that is for myocardial (7Barhanin J. Lesage F. Guillemare E. Fink M. Lazdunski M. Romey G. Nature. 1996; 384: 78-80Crossref PubMed Scopus (1392) Google Scholar, 8Sanguinetti M. C. Curran M. E. Zou A. Shen J. Spector P. S. Atkinson D. L. Keating M. T. Nature. 1996; 384: 80-83Crossref PubMed Scopus (1515) Google Scholar) and is involved in cardiac arrhythmias (1Wang Q. Curran M. E. Splawski I. Burn T. C. Millholland J. M. VanRaay T. J. Shen J. Timothy K. W. Vincent G. M. de Jager T. Schwartz P. J. Toubin J. A. Moss A. J. Atkinson D. L. Landes G. M. Connors T. D. Keating M. T. Nat. Genet. 1996; 12: 17-23Crossref PubMed Scopus (1496) Google G. W. Sesti F. Splawski I. Buck M. E. Lehmann M. H. Timothy K. W. Keating M. T. Goldstein S. A. N. Cell. 1999; 97: 175-187Abstract Full Text Full Text PDF PubMed Scopus (1173) Google Scholar, 15Sesti F. Abbott G. W. Wei J. Murray K. T. Saksena S. Schwartz P. J. Priori S. G. Roden D. M. George Jr. , A. L. Goldstein S. A. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10613-10618Crossref PubMed Scopus (442) Google Scholar). of heteromultimeric ion channel complexes other of the KCNQ and KCNE gene families may other important currents involved in and The of channel complexes in is by the of forms of epilepsy (3Charlier C. Singh N. A. Ryan S. G. Lewis T. B. Reus B. E. Leach R. J. Leppert M. Nat. Genet. 1998; 18: 53-55Crossref PubMed Scopus (817) Google Scholar, 4Singh N. A. Charlier C. Stauffer D. DuPont B. R. Leach R. J. Melis R. Ronen G. M. Bjerre I. Quattlebaum T. Murphy J. V. McHarg M. L. Gagnon D. Rosales T. O. Peiffer A. Anderson V. E. Leppert M. Nat. Genet. 1998; 18: 25-29Crossref PubMed Scopus (1026) Google Scholar), deafness (5Kubisch C. Schroeder B. C. Friedrich T. Lutjohann B. El-Amraoui A. Marlin S. Petit C. Jentsch T. J. Cell. 1999; 96: 437-446Abstract Full Text Full Text PDF PubMed Scopus (679) Google Scholar), and (16Abbott G. W. Butler M. H. Bendahhou S. Dalakas M. C. Ptacek L. J. Goldstein S. A. N. Cell. 2001; 104: 217-231Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar) by in of the two gene The structural basis for subunit-subunit interactions in fundamental for their physiological and The that of the KCNE family of potassium channel subunits may have a structural in permeation pathway has been on the basis of a of by Tai and Goldstein (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar). These investigators that Cd2+ when channels are formed with minK cysteine mutants through a direct A line of this was that the effect of Cd2+ on mutant channels could be with a potassium channel pore recently Kurokawa et al. (22Kurokawa J. Motoike H. K. Kass R. S. J. Gen. Physiol. 2001; 117: 43-52Crossref PubMed Scopus (42) Google Scholar) have shown that the effect of Cd2+ on channel complexes containing or is suggesting that minK may reside in the IKs complex. The results from Cd2+ analysis provide a framework for the location and orientation of minK within the IKs channel complex. Our data indicate that KvLQT1 residue Cys-331 forms a Cd2+ binding site in conjunction with cysteines in that the KvLQT1 S6 segment within of the minK transmembrane domain at the level of and In addition, both minK mutants and Cd2+ with KvLQT1 we the orientation of the minK transmembrane domain in relation to the KvLQT1 S6 segment Our data predict that the of the minK containing and the region of the KvLQT1 S6 segment containing residue Cys-331 with the sulfhydryl side chain between the two minK On the basis of the of Cys-331 in a KvLQT1 homology we predict that this residue projects away from the central axis of the permeation suggesting that minK line the pore of the IKs channel complex with this S6 residues to amino acids to be through the pore potassium channels Y. M. M. E. G. 1997; Full Text Full Text PDF PubMed Scopus Google Scholar, M. S. R. Nature. 2001; PubMed Scopus Google Scholar) reside on the helical Cys-331 For minK to be close to S6, we predict that an is formed between the KvLQT1 S6 segment and other transmembrane which to minK in the assembled IKs channel complex. The effect of Cd2+ on complexes is for and single channel current (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar). these of Cd2+-sensitive channels be by a direct We that its effect through an allosteric mechanism through it that an must be formed between the minK transmembrane domain and the KvLQT1 S6 Cd2+ to its binding Once Cd2+ may a on the of the S6 transmembrane to a within the channel complex that is to the pore that pore are to the spatial relationships of for binding J. M. A. J. M. R. 1998; PubMed Scopus Google Scholar), in and single channel current Furthermore, the predicted site is within the and within the transmembrane Cd2+ binding may also be with (18Tai K. K. Goldstein S. A. N. Nature. 1998; 391: 605-608Crossref PubMed Scopus (110) Google Scholar). In results provide the location and orientation of minK within the IKs channel complex. We further that Cd2+ exerts its effect on IKs through an allosteric mechanism as to direct pore blockade. These findings provide new to the structural basis of subunit-subunit interactions in voltage-gated potassium channels and to the of KCNQ/KCNE channel We and for DNA and