The S4-S5 Linker Directly Couples Voltage Sensor Movement to the Activation Gate in the Human Ether-á-go-go-related Gene (hERG) K+ Channel

Tania Ferrer, Jason Rupp, David R. Piper, Martin Tristani‐Firouzi

Journal of Biological Chemistry · 2006 · 105 citations · 29 references

DOIFull text

Open access

Concepts

Abstract

A key unresolved question regarding the basic function of voltage-gated ion channels is how movement of the voltage sensor is coupled to channel opening. We previously proposed that the S4-S5 linker couples voltage sensor movement to the S6 domain in the human ether-a'-go-go-related gene (hERG) K+ channel. The recently solved crystal structure of the voltage-gated Kv1.2 channel reveals that the S4-S5 linker is the structural link between the voltage sensing and pore domains. In this study, we used chimeras constructed from hERG and ether-a'-go-go (EAG) channels to identify interactions between residues in the S4-S5 linker and S6 domain that were critical for stabilizing the channel in a closed state. To verify the spatial proximity of these regions, we introduced cysteines in the S4-S5 linker and at the C-terminal end of the S6 domain and then probed for the effect of oxidation. The D540C-L666C channel current decreased in an oxidizing environment in a state-dependent manner consistent with formation of a disulfide bond that locked the channel in a closed state. Disulfide bond formation also restricted movement of the voltage sensor, as measured by gating currents. Taken together, these data confirm that the S4-S5 linker directly couples voltage sensor movement to the activation gate. Moreover, rather than functioning simply as a mechanical lever, these findings imply that specific interactions between the S4-S5 linker and the activation gate stabilize the closed channel conformation. A key unresolved question regarding the basic function of voltage-gated ion channels is how movement of the voltage sensor is coupled to channel opening. We previously proposed that the S4-S5 linker couples voltage sensor movement to the S6 domain in the human ether-a'-go-go-related gene (hERG) K+ channel. The recently solved crystal structure of the voltage-gated Kv1.2 channel reveals that the S4-S5 linker is the structural link between the voltage sensing and pore domains. In this study, we used chimeras constructed from hERG and ether-a'-go-go (EAG) channels to identify interactions between residues in the S4-S5 linker and S6 domain that were critical for stabilizing the channel in a closed state. To verify the spatial proximity of these regions, we introduced cysteines in the S4-S5 linker and at the C-terminal end of the S6 domain and then probed for the effect of oxidation. The D540C-L666C channel current decreased in an oxidizing environment in a state-dependent manner consistent with formation of a disulfide bond that locked the channel in a closed state. Disulfide bond formation also restricted movement of the voltage sensor, as measured by gating currents. Taken together, these data confirm that the S4-S5 linker directly couples voltage sensor movement to the activation gate. Moreover, rather than functioning simply as a mechanical lever, these findings imply that specific interactions between the S4-S5 linker and the activation gate stabilize the closed channel conformation. A fundamental property of all voltage-gated ion channels is the ability to open or close in response to changes in membrane potential. In voltage-gated K+ channels, the transmembrane domains are partitioned into distinct functional modules: a voltage-sensing module (S1-S4) and an ion-conducting module (S5-S6). Changes in the transmembrane electrical field exert a force on the highly charged S4 domain to initiate a process that culminates in the opening of the activation gate (1Bezanilla F. Physiol. Rev. 2000; 80: 555-592Crossref PubMed Scopus (720) Google Scholar). The activation gate is formed by crisscrossing of the C-terminal portions of the S6 α-helices to form a narrow aperture near the cytoplasmic interface (2Doyle D.A. Morais Cabral J. Pfuetzner R.A. Kuo A. Gulbis J.M. Cohen S.L. Chait B.T. MacKinnon R. Science. 1998; 280: 69-77Crossref PubMed Scopus (5770) Google Scholar). Channel opening is proposed to involve splaying of the S6 helices at a conserved glycine, thereby widening the aperture to allow passage of ions (3Jiang Y. Lee A. Cadene M. Chait B.T. MacKinnon R. Nature. 2002; 417: 523-526Crossref PubMed Scopus (1082) Google Scholar, 4Jiang Y. Lee A. Chen J. Cadene M. Chait B.T. MacKinnon R. Nature. 2002; 417: 515-522Crossref PubMed Scopus (1212) Google Scholar). Although substantial progress has been made in defining the structural basis of the voltage sensor and activation gate, the mechanism of coupling voltage sensing to channel opening, “electromechanical coupling” (5Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 903-908Crossref PubMed Scopus (822) Google Scholar), remains poorly defined. Electromechanical coupling may involve global rearrangements between large domains within the channel complex, for example, outward movement of S4 coupled to rearrangements in S5 that are transmitted to S6. Alternatively, the coupling mechanism might involve discreet interactions between specific residues in more localized regions. The intracellular S4-S5 linker is ideally suited to function as an electromechanical coupler given that it is physically tethered to the S4 and, as such, could function to transduce voltage sensor movement. Mutations in the S4-S5 linker of a variety of K+ channels influence activation and gating charge movement (6McCormack K. Tanouye M.A. Iverson L.E. Lin J.W. Ramaswami M. McCormack T. Campanelli J.T. Mathew M.K. Rudy B. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 2931-2935Crossref PubMed Scopus (172) Google Scholar, 7Shieh C.-C. Klemic K.G. Kirsch G.E. J. Gen. Physiol. 1997; 109: 767-778Crossref PubMed Scopus (65) Google Scholar, 8Sanguinetti M.C. Xu Q.P. J. Physiol. (Lond.). 1999; 514: 667-675Crossref Scopus (142) Google Scholar, 9Isacoff E.Y. Jan Y.N. Jan L.Y. Nature. 1991; 353: 86-90Crossref PubMed Scopus (277) Google Scholar, 10Slesinger P.A. Jan Y.N. Jan L.Y. Neuron. 1993; 11: 739-749Abstract Full Text PDF PubMed Scopus (109) Google Scholar). Hyperpolarization-activated cyclic nucleotide-gated channels with a mutation in the S4-S5 linker behave as though the link between voltage sensing, and channel opening was severed (11Chen J. Mitcheson J. Tristani-Firouzi M. Lin M. Sanguinetti M. Proc. Natl. Acad. Sci. (U. S. A.). 2001; 98: 11277-11282Crossref PubMed Scopus (133) Google Scholar). A direct interaction between the S4-S5 linker and C-terminal portion of the S6 domain (C-S6) was suggested by studies designed to impart voltage sensitivity to the voltage-independent KcsA channel. A chimera containing the pore module of KcsA and the voltage-sensing module of Shaker gated in a voltage-dependent manner only if S4-S5 linker and a 9-amino-acid stretch of C-S6 residues from Shaker were included in the chimera (12Lu Z. Klem A.M. Ramu Y. Nature. 2001; 413: 809-813Crossref PubMed Scopus (274) Google Scholar, 13Lu Z. Klem A.M. Ramu Y. J. Gen. Physiol. 2002; 120: 663-676Crossref PubMed Scopus (262) Google Scholar). We proposed that an interaction between the S4-S5 linker and C-S6 was also important in stabilizing the closed conformation of the hERG 2The abbreviations used are: hERG, human ether-a-go-go-related gene; bEAG, bovine ether-a'-go-go; HP, holding potential; Ig, gating current; IgON, on gating current; IgOFF, off gating current; tbHO2, tert-butylhydroperoxide; DTT, dithiothreitol; WT, wild type; MES, 4-morpholineethanesulfonic acid. (human ether-a'-go-go-related gene (14Warmke J.W. Ganetzky B. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 3438-3442Crossref PubMed Scopus (866) Google Scholar)) K+ channel (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). A charge reversal of a single residue in the S4-S5 linker (D540K) destabilized the closed state of hERG channels via an electrostatic interaction with an arginine in C-S6 (Arg-665). Most recently, a mechanism to explain electromechanical coupling was proposed based on the crystal structure of the mammalian voltagegated Kv1.2 channel solved in the open state (5Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 903-908Crossref PubMed Scopus (822) Google Scholar, 16Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 897-903Crossref PubMed Scopus (1858) Google Scholar). The S4-S5 linker forms an amphipathic α-helix running parallel to the cytoplasmic interface and is positioned over the C-S6 from the same subunit. Inward movement of the voltage sensor with membrane hyperpolarization is postulated to compress the S4-S5 linker against the S6 helices, resulting in closure of the activation gate (5Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 903-908Crossref PubMed Scopus (822) Google Scholar). To date, a direct physical interaction between the S4-S5 linker and the activation gate has not been established unequivocally. In this study, we used chimeras constructed from hERG and the related bovine ether-a'-go-go (bEAG) channel to identify residues in the S4-S5 linker and C-S6 that were important for normal channel closure. We introduced cysteines into these regions and demonstrated that D540C-L666C current was decreased by oxidation in a state-dependent manner associated with reduced gating current. These data suggest that electromechanical coupling is transduced via the S4-S5 linker. Moreover, rather than functioning simply as a mechanical lever, our study suggests a stringent requirement for specific residues in the S4-S5 linker and the activation gate to stabilize the channel in a closed conformation. Molecular Biology—Site-directed mutagenesis, in vitro synthesis, and injection of cRNA into Xenopus laevis oocytes were performed as described previously (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Oocytes were injected with 2-15 ng of WT or mutant hERG cRNA and incubated for 3 days as described (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). The HBS6 chimera was a kind gift of E. Ficker. Electrophysiology—Ionic currents were measured with the two-microelectrode voltage clamp technique using a GeneClamp 500 amplifier (Molecular Devices, Union City, CA). Voltage commands were generated using pCLAMP8 software (Molecular Devices), a personal computer, and a DigiData 1320 interface (Molecular Devices). Recordings were performed at room temperature (22-24 °C). Currents were filtered at 1 kHz and digitized at 2 kHz. Microelectrodes were pulled from borosilicate glass capillary tubes and had resistances of 0.8-1.5 megaohms when filled with 3 m KCl. For hERG-bEAG chimera recordings, the extracellular solution contained (in mm): 96 NaCl, 4 KCl, 1 CaCl2, 1 MgCl2, 10 HEPES, pH 7.6. For hERG Cys mutant recordings, the extracellular solution contained (in mm): 96 NaCl, 2 KCl, 1 CaCl2,2MgCl2, 10 HEPES, pH 7.6. The recording chamber was perfused at a rate of ∼1 ml/min. Gating currents were measured at room temperature using the cutopen Vaseline gap (17Stefani E. Bezanilla F. Conn P.N. Methods in Enzymology. 293. Academic Press, San Diego, CA1998: 300-318Google Scholar) recording technique and a CA-1B amplifier (Dagan Corp., Minneapolis, MN). Recordings were performed at room temperature. Signals were filtered at 5 kHz with an eight-pole Bessel filter and acquired at 20 kHz. Microelectrodes were filled with 3 m KCl to obtain resistances of 0.1-0.2 megaohms. The extracellular solution (top and guard chambers) consisted of (in mm): 120 tetraethylammonium-MES, 2 Ca-MES, 10 HEPES, pH 7.4, and the intracellular solution consisted of (in mm): 120 tetraethylammonium-MES, 10 HEPES, pH 7.4. The oocyte was permeabilized by adding 0.2% saponin to the lower chamber for 2 min. Intracellular K+ was depleted by clamping the membrane to 0 mV for 30 min. To ensure complete block of hERG channels, the specific hERG blocker MK-499 (50 μm) was added to the external and internal solutions after depletion of intracellular K+ as described (18Piper D.R. Varghese A. Sanguinetti M.C. Tristani-Firouzi M. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 10534-10539Crossref PubMed Scopus (99) Google Scholar). Linear leak and capacitance currents were partially compensated by analog circuitry and subtracted on-line using a p/-8 protocol. (19Armstrong C.M. Bezanilla F. J. Gen. Physiol. 1977; 70: 567-590Crossref PubMed Scopus (766) Google Scholar) from a pre-pulse potential of -140 or -160 mV and a holding potential (HP) of -110 mV. Non-linear transient currents were not detected in uninjected oocytes, indicating that gating currents detected in hERG-injected oocytes were not contaminated by constitutively expressed channels. Data Analysis—The isochronal voltage dependence of the activation of WT and mutant hERG channels was determined from tail current amplitudes measured at -70 or -120 mV following 2-s depolarizing voltage steps that were applied from an HP of -90 or -120 mV as described (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Tail current was potential and to a function using software is the voltage at the currents are and is the the tail current data are expressed as of S6 into hERG Channel E. J. A. A.M. 1998; PubMed Scopus Google Scholar) that of the S6 domain the 5 residues of the of into hERG activation gating that channels to constitutively open at is an given the between the S6 domains of these channels and suggests a stringent requirement for specific residues to stabilize the activation gate in closed conformation. WT hERG and HBS6 currents by voltage steps to between and mV from an HP of -120 mV are in A and B. hyperpolarization not the opening of WT hERG channels WT hERG currents. tail currents that of channels that were by the voltage by HBS6 channels were open at HP of -120 as by an holding current that was by the of the hERG channel blocker and depolarizing voltage steps and outward currents that an channel and to -120 mV tail currents that the of channels from by and channel HBS6 channels partially a at Voltage steps to as as mV to complete channel closure The HBS6 activation that a of channels at mV In the HBS6 activation was by mV when with WT hERG the the S6 of was to a closed conformation when into the hERG channel Channel by C-terminal S6 the residues for the closed conformation of the HBS6 we C-terminal HBS6 residues to the to hERG chimera to channel as by the holding current at -120 mV 2 residues at the end of the S6 domain chimera channel closure of S6 residues chimera in a channel that more to WT hERG between the S4-S5 and C-terminal S6 Channel channel closure was by residues in the S6 we the effect of only the 2 of the S6 domain and in the of the HBS6 chimera Channel closure was in the chimera A and We that these channels were to close the chimera contained hERG in the S4-S5 linker and that regions were for hERG between the S4-S5 linker and the activation gate might explain the channel closure in the HBS6 To this we the residues in the HBS6 S4-S5 linker to the that S4-S5 linker and S6 The S4-S5 linker of hERG from at and The and in the of the HBS6 chimera also channel closure and between the of the S4-S5 linker and C-terminal S6 domain destabilized the closed conformation of the HBS6 a is an given the between these related channels and that specific interactions between these regions are to the activation gate. between Cys in the S4-S5 and C-terminal S6 previously proposed that an electrostatic interaction between and the closed conformation of hERG channels (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). In to the HBS6 chimera studies as a critical residue for stabilizing the closed conformation. To and are within we introduced cysteines at these and probed the effect of oxidation on current We that formation of a disulfide bond between these regions the channel in a single state. The oxidizing was used in this study the residues of are near the intracellular portion of the were 10 to mV for 2 to and channels and then to -70 mV to tail currents of the tail 2 was perfused the voltage protocol. of were with normal extracellular in WT hERG tail current decreased with current in the of current decreased D540C-L666C current decreased an effect that was not by with normal extracellular solution Tail currents from the single Cys mutant decreased with that the in current with D540C-L666C was The in D540C-L666C current by that formation of a disulfide bond between the S4-S5 linker and C-S6 locked the channel in a state. To the disulfide bond could with a D540C-L666C hERG were to by 20 or normal extracellular solution partially the effect of by current to of when with for perfused with normal external solution To the state dependence of the effect on D540C-L666C channels, were at -90 mV or mV the of 2 D540C-L666C channels were in the closed state and to tbHO2, current was reduced following a and D540C-L666C channels in the state the channels from and These that of and are in the closed state and from in the Cys between the S4-S5 and C-terminal S6 of the Voltage and form a disulfide bond in the closed state and channel opening, might that voltage sensor movement by oxidation. To this we measured gating currents and after was by a voltage to mV from a potential of -140 mV for D540C-L666C channels and a voltage to mV from a potential of -160 mV for we charge reversal or of the rate of hERG channel activation M.C. Xu Q.P. J. Physiol. (Lond.). 1999; 514: 667-675Crossref Scopus (142) Google Scholar, D.R. Sanguinetti M.C. Tristani-Firouzi M. J. Biol. Chem. 2005; 280: Full Text Full Text PDF PubMed Scopus Google Scholar), the in the and the rate of and D.R. Sanguinetti M.C. Tristani-Firouzi M. J. Biol. Chem. 2005; 280: Full Text Full Text PDF PubMed Scopus Google Scholar). the of and were also for D540C-L666C and channels an effect to mutation of were after 2 was applied to the external solution (top and guard chambers) for were 2-s to mV. The of reduced D540C-L666C Ig, with effect on The of at mV was of for D540C-L666C and for WT hERG was by not Taken with the chimera the that disulfide bond formation D540C-L666C suggests that the S4-S5 linker couples movement of the S4 voltage sensor directly to the activation gate in S6. Although has been defining the structural basis of the voltage sensor and the activation gate, the mechanism of electromechanical coupling in voltage-gated ion channels remains We and proposed that the cytoplasmic S4-S5 linker voltage sensor movement to the activation gate in the C-terminal S6 Z. Klem A.M. Ramu Y. J. Gen. Physiol. 2002; 120: 663-676Crossref PubMed Scopus (262) Google Scholar, M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). of the crystal structure of the voltage-gated Kv1.2 channel that the S4-S5 linker and the C-terminal S6 domain are within proximity S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 897-903Crossref PubMed Scopus (1858) Google Scholar). on the crystal structure of Mackinnon and (5Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 903-908Crossref PubMed Scopus (822) Google Scholar) proposed that the S4-S5 linker as a that a mechanical force on the activation gate, by the electrical force of the transmembrane potential. findings imply that are stringent for the S4-S5 linker to function as a mechanical to stabilize the closed channel conformation. between the of the S4-S5 linker and the end of S6 in the HBS6 chimera the of the that passage of ions was at at the activation gate is Mutations in the activation gate of Shaker also channel resulting in a to that of the HBS6 of the in the of Shaker in a at and voltage-dependent opening at J. Gen. Physiol. 2002; PubMed Scopus (150) Google Scholar, M. J. Gen. Physiol. 2003; PubMed Scopus Google Scholar). The of these was of voltage sensor movement M. J. Gen. Physiol. 2003; PubMed Scopus Google Scholar). channel at poorly between consistent with of the activation gate. The described the effect of the as an in the between the closed and open M. J. Gen. Physiol. 2003; PubMed Scopus Google Scholar). the voltage of the at in the HBS6 chimera that ion of movement of the voltage the of the activation gate with to the S4-S5 linker is that the activation gate a at channel closure is when the S4-S5 linker and C-terminal S6 are between the S4-S5 linker and the S6 in the chimera also channel of voltage sensor movement Z. Klem A.M. Ramu Y. J. Gen. Physiol. 2002; 120: 663-676Crossref PubMed Scopus (262) Google Scholar). Moreover, Shaker in these regions was for voltage-dependent gating (12Lu Z. Klem A.M. Ramu Y. Nature. 2001; 413: 809-813Crossref PubMed Scopus (274) Google Scholar, 13Lu Z. Klem A.M. Ramu Y. J. Gen. Physiol. 2002; 120: 663-676Crossref PubMed Scopus (262) Google Scholar). Taken together, these studies the that the S4-S5 linker and the C-terminal S6 to the activation gate in closed conformation. findings suggest that and form a disulfide bond oxidizing to the channel in the closed state and movement of the voltage In the of disulfide bond formation between these 2 residues is suggested by the state-dependent of D540C-L666C channels not the single Cys or and the ability to with the of a disulfide bond between 2 Cys that the are within 5 and J. Biol. PubMed Scopus Google Scholar, and Molecular and Scholar). Although we the that or with an our data are consistent with the that and are within 5 in the closed state and from in the The of the in Kv1.2 and are by in the open channel state consistent with the that membrane D540C-L666C channels from near the end of the S4 to a to the end of the S6 in the closed proposed by Mackinnon and (5Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 903-908Crossref PubMed Scopus (822) Google Scholar), movement of the S4 by membrane force the S4-S5 linker against the C-terminal end of the aperture of the activation gate and near Although this a large of the S4-S5 linker with to S6 rearrangements of to been in membrane by disulfide J. Biol. PubMed Scopus Google Scholar). of the S4-S5 linker against the S6 domain could as the of of the proposed of S4 the (1Bezanilla F. Physiol. Rev. 2000; 80: 555-592Crossref PubMed Scopus (720) Google Scholar, Rev. PubMed Google Scholar), movement Y. Chen J. Lee A. MacKinnon R. Nature. 2003; PubMed Scopus Google Scholar), or of the S4 B. R. B. Bezanilla F. Nature. 2005; PubMed Scopus Google Scholar). The of between and suggests that the between these residues in the closed state 5 or that the of the is for disulfide bond formation J. Biol. PubMed Scopus Google Scholar). We the given our of the S4-S5 linker and C-terminal S6 domain in The charge mutation destabilized the closed hERG channel channel with membrane closed state was an electrostatic interaction between residues in and (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). of channel charge reversal is not or of (15Tristani-Firouzi M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar), a for this residue in the of the closed state of hERG channels. of the C-S6 residues in Shaker the of this in coupling to the voltage sensor S. R. J. Gen. Physiol. 2002; 120: PubMed Scopus Google Scholar, S. R. J. 2003; Full Text Full Text PDF PubMed Scopus Google Scholar). These not a for regions of the channel in coupling voltage sensing to channel opening. Although a for the S4-S5 linker in electromechanical coupling has been previously proposed (5Long S.B. Campbell E.B. Mackinnon R. Science. 2005; 309: 903-908Crossref PubMed Scopus (822) Google Scholar, K. Tanouye M.A. Iverson L.E. Lin J.W. Ramaswami M. McCormack T. Campanelli J.T. Mathew M.K. Rudy B. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 2931-2935Crossref PubMed Scopus (172) Google Scholar, 10Slesinger P.A. Jan Y.N. Jan L.Y. Neuron. 1993; 11: 739-749Abstract Full Text PDF PubMed Scopus (109) Google Scholar, 13Lu Z. Klem A.M. Ramu Y. J. Gen. Physiol. 2002; 120: 663-676Crossref PubMed Scopus (262) Google Scholar, M. Chen J. Sanguinetti M.C. J. Biol. Chem. 2002; 277: 18994-19000Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar, R. 2000; PubMed Scopus Google Scholar), our data a direct interaction between the S4-S5 linker and the activation gate in the process of coupling voltage sensing to channel opening. In to voltage sensor specific residues in the S4-S5 linker stabilize the activation gate in the closed conformation. These with the recently solved crystal structure of the S4-S5 linker as the structural link that couples voltage sensing to channel opening. We Sanguinetti for and J. and for

References

29