HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 21, 18994-19000, May 24, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/21/18994    most recent M200410200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tristani-Firouzi, M. Articles by Sanguinetti, M. C. Search for Related Content PubMed PubMed Citation Articles by Tristani-Firouzi, M. Articles by Sanguinetti, M. C. Social Bookmarking                      What's this?

Interactions between S4-S5 Linker and S6 Transmembrane Domain Modulate Gating of HERG K⁺ Channels^*

Martin Tristani-Firouzi^§, Jun Chen^¶, and Michael C. Sanguinetti^¶

From the  Department of Pediatrics, ^¶ Department of Medicine, and  Eccles Program in Human Molecular Biology and Genetics, University of Utah School of Medicine, Salt Lake City, Utah 84112

Received for publication, January 15, 2002, and in revised form, February 20, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Outward movement of the voltage sensor is coupled to activation in voltage-gated ion channels; however, the precise mechanism and structural basis of this gating event are poorly understood. Potential insight into the coupling mechanism was provided by our previous finding that mutation to Lys of a single residue (Asp⁵⁴⁰) located in the S4-S5 linker endowed HERG (human ether-a-go-go-related gene) K⁺ channels with the unusual ability to open in response to membrane depolarization and hyperpolarization in a voltage-dependent manner. We hypothesized that the unusual hyperpolarization-induced gating occurred through an interaction between Lys⁵⁴⁰ and the C-terminal end of the S6 domain, the region proposed to form the activation gate. Therefore, we mutated six residues located in this region of S6 (Ile⁶⁶²-Tyr⁶⁶⁷) to Ala in D540K HERG channels. Mutation of Arg⁶⁶⁵, but not the other five residues, prevented hyperpolarization-dependent reopening of D540K HERG channels. Mutation of Arg⁶⁶⁵ to Gln or Asp also prevented reopening. In addition, D540R and D540K/R665K HERG reopened in response to hyperpolarization. Together these findings suggest that a single residue (Arg⁶⁶⁵) in the S6 domain interacts with Lys⁵⁴⁰ by electrostatic repulsion to couple voltage sensing to hyperpolarization-dependent opening of D540K HERG K⁺ channels. Moreover, our findings suggest that the C-terminal ends of S4 and S6 are in close proximity at hyperpolarized membrane potentials.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The human ether-a-go-go-related gene (HERG)¹ encodes the -subunit of the ion channel underlying the cardiac delayed rectifier K⁺ current, I_Kr (1, 2). Similar to other voltage-gated K⁺ (Kv) channels, HERG channels are closed at negative transmembrane potentials and open or inactivate in response to membrane depolarization. Membrane depolarization causes outward displacement of the voltage-sensing S4 domain (3-6), which precedes opening of the activation gate. Although the structural basis of the activation gate is not clear, substantial evidence implicates a critical role for the S6 -helices (7-10). The equivalent domains of the KcsA channel (inner helices) line the channel pore and crisscross near the cytoplasmic side of the membrane to create a narrow aperture (7). Narrowing or widening of the aperture is hypothesized to mediate channel closure or opening, respectively. Activation of KcsA channels by protons involves counterclockwise rotation of the inner helices, which presumably widens the diameter of the aperture to allow ion permeation (10). Substituted cysteine accessibility mutagenesis experiments have identified specific residues in the S6 domain of Shaker channels that are located on either side of the putative activation gate (8).

What is the structural basis for coupling the movement of the voltage sensors with opening of the activation gate? Displacement of S4 could be transmitted to the activation gate via the concerted movement of domains S4-S6 or via specific interactions between more limited regions of the channel. For example, the intracellular S4-S5 linker may provide a direct link between the S4 domain and the activation gate (9, 11). Voltage-dependent modification of the S4-S5 linker by thiol-reactive reagents suggests that conformational changes in the linker are coupled to movement of the S4 domain (12). Mutations in the S4-S5 linker influence the activation gating properties of Kv2.1 and Kv3.1 channels (13). Finally, a point mutation located in the S4-S5 linker (D540K) fundamentally alters the gating properties of HERG channels (14). D540K HERG channels have the unique ability to open in response to membrane hyperpolarization while retaining the ability to activate and inactivate in response to membrane depolarization. Although these studies suggest an important role for the S4-S5 linker in channel gating, specific interactions between the linker and the activation gate have not been demonstrated.

The unique gating properties of D540K HERG provide the opportunity to characterize specific interactions that mediate coupling between voltage sensing and channel activation. We hypothesized that the closed state of D540K HERG channels is destabilized by the interaction of Lys⁵⁴⁰ with a specific amino acid(s) of the S6 domain. In this study we focused our attention on the C-terminal region of S6 because of the proposed role of the homologous region in KcsA channel activation. Six adjacent residues in S6 of D540K HERG were individually mutated to Ala, and the ability of the channels to open in response to membrane hyperpolarization was assessed. This approach identified a single residue (Arg⁶⁶⁵) that interacted with Lys⁵⁴⁰ to permit hyperpolarization-induced channel opening. Further mutagenesis suggested that the mechanism of this unique gating was electrostatic repulsion between basic residues at positions 540 and 665.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Molecular Biology-- Mutations were introduced into wild-type (WT) HERG using site-directed mutagenesis as previously described (15). Before use in expression experiments, the constructs were characterized by restriction mapping and DNA sequence analyses. cRNA for injection into oocytes was prepared with SP6 Capscribe (Roche Molecular Biochemicals) following linearization with EcoRI.

Injection of RNA and Voltage Clamp of Oocytes-- Xenopus laevis frogs were anesthetized by immersion in 0.2% tricaine for 15 min. Ovarian lobes were digested with 2 mg/ml Type 1A collagenase (Sigma) in Ca²⁺-free ND96 solution for 1.5 h to remove follicle cells. Stage IV and V oocytes were injected with WT or mutant HERG cRNA (5-15 ng) and then cultured in Barth's solution supplemented with 50 µg/ml gentamycin and 1 mM pyruvate at 18 °C. Barth's solution contained (in mM): 88 NaCl, 1 KCl, 0.4 CaCl₂, 0.33 Ca(NO₃)₂, 1 MgSO₄, 2.4 NaHCO₃, 10 HEPES (pH 7.4).

The two-microelectrode voltage clamp technique (16) was used to record membrane currents in oocytes 1-3 days after cRNA injection. To attenuate endogenous chloride currents, Cl was replaced with MES in the external solution containing (in mM) 96 NaMES, 2 KMES, 2 CaMES₂, 5 HEPES, 1 MgCl₂, adjusted to pH 7.6 with methanesulfonic acid. Currents were recorded at room temperature (21-23 °C). Glass microelectrodes were filled with 3 M KCl, and their tips were broken to obtain tip resistances of 0.8-1.5 megaohms. Oocytes were voltage-clamped with a GeneClamp 500 amplifier (Axon Instruments, Foster City, CA). Voltage commands were generated using pClamp8 software (Axon Instruments), a personal computer, and a DigiData 1200 series interface (Axon Instruments). The oocyte membrane potential was held at 80 mV between test pulses, unless noted otherwise.

Data Analysis-- The isochronal voltage dependence of activation of the hyperpolarization-induced open state (O_h) and the normal depolarization-induced open state (O) of D540K HERG channels were determined from tail currents measured at 70 mV following 2-s hyperpolarizing or depolarizing voltage steps, respectively. Tail current amplitude (I_t) was determined by fitting the decaying portion of tail currents to a bi-exponential function and extrapolating to the beginning of the repolarizing step. I_t was plotted versus test potential (V_t) and fitted to the sum of two Boltzmann distributions using ORIGIN software (Northampton, MA).

(Eq. 1)

V_1/2Oh and V_1/2O are the voltages at which the O_h and O state currents are half-activated. k_Oh and k_O are the slope factors for the biphasic activation curve. I_maxOh and I_maxO are the maximum O_h and O state tail current values. For this analysis we assume that the relative probability for the O_h and O state approaches zero at extreme voltages. For WT HERG (no O_h state) the first half of Equation 1 was set to zero. The time course of deactivation was obtained by fitting decaying tail currents to a bi-exponential function. Data are expressed as the means ± S.E. (n = number of oocytes).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Biophysical Properties of WT and D540K HERG-- The properties of WT and D540K HERG have been described previously (1). Briefly, WT HERG current activated very slowly in response to weak depolarizations (e.g. 20 mV), whereas the activation rate was relatively fast at stronger depolarizations (e.g. +20 mV) (Fig. 1A, left panel). Fast inactivation predominated at depolarized potentials, resulting in a decrease in steady state current magnitude at potentials positive to 10 mV (Fig. 1B). Membrane hyperpolarization to potentials negative to 90 mV failed to induce inward current (Fig. 1, A and B). Repolarization to 70 mV after a depolarizing test pulse induced a tail current with a rapid rising phase followed by a slow decay, as channels that were inactivated during the test pulse rapidly recovered from inactivation and then slowly deactivated (Fig. 1A, left panel).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Comparison of WT and D540K HERG K+ currents recorded from Xenopus oocytes. A, representative currents elicited from a holding potential of 80 mV. Two-second pulses were applied in 20-mV increments to potentials ranging from 160 to +40 mV. D540K HERG tail currents are also shown on an expanded scale (right panel, inset). B, current-voltage relationships for WT HERG (n = 8) and D540K HERG (n = 7). C, voltage-dependent activation of D540K HERG channel current. Currents were normalized to the peak tail current measured after a 2-s pulse to 160 mV. Data were fit with the sum of two Boltzmann functions. For depolarization-dependent channel opening, V1/2 = 15.5 ± 2.5 mV, and the slope factor was 12.4 ± 0.4 mV. For hyperpolarization-dependent channel opening, V1/2 = 109.8 ± 1.8 mV, and the slope factor was 11.7 ± 0.9 mV.

As reported previously (14), D540K HERG current activated nearly instantaneously upon depolarization from a holding potential of 80 mV. Hyperpolarization to potentials negative to 90 mV induced a slowly activating current that attained a steady state amplitude in 2 s (Fig. 1A, right panel). The resulting current-voltage (I-V) relationship showed that inward current magnitude was much greater than outward current for equivalent electrochemical driving forces (Fig. 1B). Repolarization to 70 mV after a hyperpolarizing test pulse (e.g. 160 mV) induced a tail current whose magnitude was larger and decay slower than tail currents induced after depolarizing test pulses (Fig. 1A, inset). Tail currents after depolarizing test pulses (e.g. +40 mV) initially increased as channels recovered from inactivation and then decayed as channels deactivated into the closed state (Fig. 1A, inset).

The voltage dependence for activation of the O_h and O states for D540K HERG was determined by plotting peak tail current at 70 mV versus test potential and fitting the relationship to the sum of two Boltzmann functions. The normalized voltage dependence of the D540K channel opening had an inverted bell-shaped relationship (Fig. 1C). The V_1/2 values for O and O_h state activation were 15.5 ± 2.5 and 109.8 ± 1.8 mV, respectively. Therefore, the voltage dependence of depolarization- and hyperpolarization-induced channel opening was sufficiently separated to result in a voltage range near 60 mV where most channels were in a closed state.

Ala Substitution in the C-terminal Region of S6-- We tested the hypothesis that the O_h state is dependent upon interactions between Lys⁵⁴⁰ and a specific residue(s) in the C-terminal region of S6. Amino acids 662-667 (Fig. 2) of D540K HERG were individually mutated to Ala, and the resulting double mutant channel function was assayed using the voltage pulse protocol illustrated in Fig. 1A. An amino acid mutation that prevented hyperpolarization-induced channel activation was considered a candidate residue for direct interaction with Lys⁵⁴⁰.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Single HERG subunit with amino acid sequence of S4-S5 linker and S6 domain in single-letter code. Residues that were mutated in the study are shown as bold letters.

The double mutant channels could be classified into three distinct groups based upon the relative magnitudes of inward and outward currents and the rate of deactivation from the O_h state. Group 1, comprising D540K/I663A, D540K/Q664A, and D540K/Y667A HERG channels, were most similar to D540K HERG. These mutant channels were characterized by large inward currents activated by hyperpolarization and extremely fast activation of outward current (Fig. 3, A-C). Unlike D540K HERG channels, Group 1 mutant channels exhibited a relatively large instantaneous current relative to the time-dependent component of inward current in response to hyperpolarization. In addition, deactivation of channels from the O_h state was slower than D540K HERG (Fig. 4). Group 2, consisting of D540K/I662A and D540K/L666A HERG channels, had relatively small hyperpolarization-activated inward currents compared with depolarization-activated outward currents (Fig. 3, D and E). Furthermore, deactivation of current after hyperpolarizing pulses was fast compared with deactivation after depolarizing pulses, and this rate was faster than D540K HERG channels (Fig. 4). In contrast to Groups 1 and 2, D540K/R665A HERG channels did not open in response to hyperpolarizing pulses as negative as 160 mV (Fig. 3F). The gating kinetics of D540K/R665A HERG channels were also different from the other double mutant channels. The rates of current activation and deactivation were much slower (Fig. 3F).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Currents recorded from double mutant HERG channels. The D540K mutation was combined with the indicated Ala mutation of a single residue in the S6 domain. Mutant channels were grouped based upon the relative magnitudes of inward and outward currents and the rate of deactivation from the Oh state. A-C (Group 1), mutation of Ile633, Gln664, or Tyr667 to Ala yielded channels that conducted large inward currents in response to hyperpolarization and slowly decaying tail currents upon repolarization to 70 mV, similar to D540K. The largest tail current in A, B, and C (indicated by arrow) was elicited in response to a preceding test pulse to 160 mV. D and E (Group 2), mutation of Ile662 or Leu666 to Ala yielded channels with small inward currents with hyperpolarization and rapidly decaying tail currents upon repolarization to 70 mV. Insets in D and E show two tail currents on an expanded time scale; test potentials (in mV) are listed. F, D540K/R665A HERG channels did not open in response to membrane hyperpolarization. Arrows indicate preceding test pulse potentials (in mV). Currents were elicited with 2-s pulses (4 s in B) applied in 20-mV increments to potentials ranging from 160 mV to +40 mV from a holding potential of 80 mV (except D where holding potential was 100 mV). The test potential is indicated in mV for the largest inward and outward currents. At potentials positive to that indicated for the largest current (e.g. more than 60 mV in A), the currents became smaller because of more intense channel inactivation.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Time constants for Oh state current deactivation at 70 mV following a pulse to 140 mV (n = 4-10).

The voltage dependence of channel activation was determined by plotting the relative amplitude of tail currents at 70 mV as a function of test potential. The currents were normalized to the largest tail current amplitude. For Group 1 channels, currents were normalized to the tail current measured after a pulse to 160 mV (Fig. 5A). For Group 2 and D540K/R665A HERG, currents were normalized to the tail current measured after a depolarizing pulse (Fig. 5B). The voltage dependence of channel activation was fit with the sum of two Boltzmann functions, one describing the hyperpolarization-dependent channel opening and another describing the depolarization-dependent channel opening. The V_1/2 for O_h state activation varied between 65 and 112 mV, and the slope factor varied between 9 and 31 mV for the different double mutant channels, compared with a V_1/2 of 110 mV and a slope factor of 9 mV for D540K (Table I). The V_1/2 for O state activation varied between 11 and 37 mV with a slope factor between 12 and 22 mV for the different double mutant channels, compared with 16 mV and 12 mV for D540K (Table I). The O state activation parameters of D540K/I663A HERG channels could not be accurately determined because the relative amplitude of tail currents was too small. D540K/R665A HERG channels did not reopen with hyperpolarization. Unlike the other mutants, the voltage dependence of O state activation for D540K/R665A HERG was best described by the sum of two Boltzmann functions (Table I).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Voltage-dependent activation of double mutant HERG channels. The D540K mutation was combined with the indicated Ala mutation of a single residue in the S6 domain. A, plot of tail currents measured at 70 mV following 2-s (or 4-s for D540K/Q664A) test pulses to the indicated voltage. Currents were normalized to the peak tail current measured following a test pulse to 160 mV. B, plot of tail currents measured at 70 mV following 2-s test pulses to the indicated voltage. Currents were normalized to the peak tail current measured following a test pulse to +20 mV for D540K/R665A HERG, +40 mV for D540K/I662A, or +60 mV for D540K/L666A HERG. The V1/2 and slope factors for the curves were determined by fitting the data to the sum of two Boltzmann functions (see Table I for values).

View this table: [in this window] [in a new window]   Table I Voltage dependence of Oh and O current activation The voltage dependence of Oh and O current activation was determined by plotting peak tail current amplitude at 70 mV versus test potential and fitting the data to the sum of two Boltzmann functions. V1/2, potential of half-maximal Oh or O state current activation; k, slope factor; n, number of oocytes. N/A (not applicable), no Oh current detected. N.D. (not determined), unable to accurately determine O current activation parameters due to small tail current amplitude.

In summary, mutation of Arg⁶⁶⁵ to Ala, but not the other five residues in the scanned region of the S6 domain, rescued WT channel function by preventing the hyperpolarization-dependent reopening of D540K HERG channels. Although D540K-I662A and D540K-L666A HERG channels reopened with membrane hyperpolarization, the relative magnitude of inward current was small and deactivation from the O_h state was rapid compared with D540K HERG, consistent with destabilization of the O_h state. These findings demonstrate that specific residues in S6 can influence the stability of the D540K HERG channel O_h state and identify a critical role for Arg⁶⁶⁵ in hyperpolarization-dependent channel gating.

Basic Residues at Positions 540 and 665 Are Required for Hyperpolarization-induced Channel Opening-- To further investigate the role of the amino acid at position 665 in mediating hyperpolarization-induced reopening of D540K HERG channels, we mutated Arg⁶⁶⁵ to an acidic residue (Asp), a polar residue (Gln), and a basic residue (Lys). Neither D540K-R665D nor D540K-R665Q channels reopened with hyperpolarizing pulses (Fig. 6, A, B, and E). In contrast, substitution of Arg⁶⁶⁵ with another basic residue, Lys, induced channels to reopen in response to hyperpolarization (Fig. 6C). The biophysical properties of D540K-R665K HERG channels were similar to D540K HERG channels, with the exceptions that depolarization-induced activation was shifted by 18 mV and the slope of the O_h activation curve was less steep (Table I).

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Hyperpolarization-dependent HERG channel opening requires basic amino acids at positions 540 and 665. A and B, D540K HERG channels do not reopen at negative potentials if R665 is mutated to an Asp (A) or Gln (B). C and D, HERG channels do reopen at negative potentials if the D540K mutation is combined with R665K (C) or if Asp540 is mutated to an Arg (D). E, current-voltage relationships for channel currents shown in panels A-D (n = 5-12).

To confirm the importance of a basic residue at position 540, we substituted another basic residue, Arg, at this position in WT HERG. Like D540K HERG, D540R HERG channels also reopened in a time- and voltage-dependent manner with membrane hyperpolarization (Fig. 6D). The V_1/2 for the O_h and O state activation curves for D540R were shifted by about +10 mV compared with D540K HERG (Table I). These findings underscore the requirement for basic residues at position 665 in S6 and position 540 in the S4-S5 linker to permit hyperpolarization-induced channel opening and suggest that electrostatic repulsion might mediate this unusual gating.

Amino Acid Substitutions of Arg⁶⁶⁵ Markedly Slow Channel Deactivation-- If an electrostatic interaction between residues 540 and 665 was the sole mechanism of hyperpolarization-induced channel opening, then substitution of Arg⁶⁶⁵ with an acidic residue in WT HERG (D540) might also produce a channel with properties similar to D540K HERG. We therefore substituted Glu or Asp for Arg at position 665 in WT HERG. R665E HERG channels did not functionally express. R665D HERG channels expressed but did not reopen with membrane hyperpolarization, even when cells were pulsed to 200 mV (data not shown). The R665D mutation did, however, affect the gating properties in that the transition from the depolarized O state to the closed state was markedly slow compared with WT HERG (Fig. 7). To further characterize the role of Arg⁶⁶⁵ in WT HERG channel gating, we studied the effects of substitution of Arg⁶⁶⁵ with the hydrophobic residue Ala or the polar residue Gln. R665A, and to a lesser extent R665Q, HERG channels deactivated more slowly than WT HERG (Fig. 7). Mutations at position 665 also shifted the voltage dependence of O state activation to more hyperpolarized potentials compared with WT or D540K HERG (R665D, V_1/2 = 35.6 ± 1.4 mV, k = 10.6 ± 0.6 mV; R665A, V_1/2 = 30.9 ± 0.6 mV, k = 6.5 ± 0.1 mV; and R665Q HERG, V_1/2 = 19.4 ± 1.2 mV, k = 7.1 ± 0.3 mV). Taken together, these data indicate that mutation of Arg⁶⁶⁵ slows deactivation, providing further evidence for the critical role of this residue in HERG channel gating.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Mutation of Arg665 slows HERG channel deactivation. Normalized tail currents recorded at 60 mV after a 1-s activating pulse to +20 mV (inset shows pulse protocol) are shown. R665D channels deactivated more slowly than R665A, R665Q, or WT HERG. The time constants for the fast component of deactivation were as follows: R665D, 1.58 ± 0.20 s; R665A, 0.86 ± 0.04 s; R665Q, 0.35 ± 0.03 s; and WT HERG, 0.14 ± 0.01 s. The time constants for the slow component of deactivation were: R665D, 12.5 ± 3.8 s; R665A, 7.21 ± 0.42 s; R665Q, 2.46 ± 0.19 s; and WT HERG, 1.23 ± 0.07 s (n = 6-9).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Despite recent advances in the understanding of the structural basis of the voltage sensor and the activation gate, little is known about how voltage sensing is coupled to channel opening. Potassium channel activation likely involves rotation of the S4 (17, 18) and S6 (10) domains. S6 rotation presumably opens the channel by increasing the diameter of the aperture formed by crisscrossing of the S6 domains (10, 19). The coupling between voltage sensing and channel opening could be mediated via a concerted movement of the S4-S6 domains or transduced by a discrete region of the protein (e.g. the S4-S5 linker) (9). The coupling mechanism is usually vectorial; membrane depolarization induces opening of the Kv channels, whereas hyperpolarization induces opening of the HCN channels. In contrast, opening of D540K HERG channels is coupled to both depolarization and hyperpolarization. The single channel conductance of the O and O_h states is identical (20), implying that D540K affects the activation gate but not the selectivity filter. A simplified gating scheme for D540K HERG channels is shown (Scheme I). In the same manner that depolarization-induced outward movement of the S4 domain is coupled to normal channel opening (3-6), we propose that the voltage dependence of the O_h state derives from hyperpolarization-induced inward movement of S4. We hypothesized that the Lys⁵⁴⁰ point mutation in HERG causes channels to reopen in response to membrane hyperpolarization because of a direct interaction with the activation gate. Therefore, we sequentially mutated to Ala residues 662-667 located in the C-terminal region of S6. Of the six Ala-substituted double mutants examined, only D540K-R665A HERG channels failed to reopen upon hyperpolarization, suggesting a direct interaction between these two residues was necessary to mediate channel reopening. Mutation of the other five S6 residues to Ala had a variable effect on the ability of the D540K HERG channel to reopen at negative potentials. Mutation of Ile⁶⁶³, Gln⁶⁶⁴, or Tyr⁶⁶⁷ (Group 1) had the least effect on channel properties compared with D540K HERG. Mutation of Ile⁶⁶² or Tyr⁶⁶⁶ (Group 2) reduced the relative magnitude of inward current activated by hyperpolarization and accelerated the rate of deactivation from the O_h state. Thus, mutation of Ile⁶⁶² or Tyr⁶⁶⁶ to Ala reduced the functional consequence of the presumed interaction between D540K and Arg⁶⁶⁵. A possible explanation for these findings is the location of the Group 2 versus the Group 1 residues relative to Arg⁶⁶⁵. Using the crystal structure of KcsA, a homology model of HERG was constructed (21). According to this model, Arg⁶⁶⁵ faces away from the central cavity of the channel. The side groups of Arg⁶⁶⁵, Ile⁶⁶², and Tyr⁶⁶⁶ are clustered together (Fig. 8). In contrast, the Group 1 residues Ile⁶⁶³, Gln⁶⁶⁴, and Tyr⁶⁶⁷ project away from Arg⁶⁶⁵. Mutation of Ile⁶⁶² or Tyr⁶⁶⁶ to the smaller Ala may alter the orientation of Arg⁶⁶⁵ in a manner that results in a diminished interaction with Lys⁵⁴⁰, whereas mutation of any one of the Group 1 amino acids has no significant effect on the Lys⁵⁴⁰-Arg⁶⁶⁵ interaction.

View larger version (8K): [in this window] [in a new window]   Scheme I.

View larger version (43K): [in this window] [in a new window]   Fig. 8.   Putative location of Arg665 and nearby residues in the S6 domain of a HERG channel subunit. The homology model of HERG pore helix-S6 domain was based on the crystal structure of the KcsA potassium channel (7). The subunit was rotated by 90o (middle panel) and 180o (right panel). The S5 and S6 domains are colored yellow, and the pore helix is colored green. The six amino acids in S6 that were mutated to Ala are shown in colored space-filling mode: Ile662 (blue); Ile663 (orange); Gln664 (cyan); Arg665 (red); Leu666 (green); Tyr667 (white). Note that the side chains of Ile662 and Leu666 face toward position Arg665, whereas the side chains Ile663, Gln664, and Tyr667 face away (see text for details).

Electrostatic Repulsion between 540 and 665 Mediates Hyperpolarization-induced Activation-- Substitution of Arg⁶⁶⁵ with a basic residue (Lys), but not an acidic (Asp) or polar residue (Gln), restored the ability of D540K HERG channels to enter the O_h state. Likewise, substitution of Asp⁵⁴⁰ with a basic amino acid, Arg (Fig. 6D), but not Ala (14) resulted in channels that were capable of hyperpolarization-induced opening. Thus, basic residues are required at position 540 in the S4-S5 linker and position 665 in the C-terminal region of S6 to enable hyperpolarization-induced channel opening. This finding suggests that electrostatic repulsion repulsion mediates this channel reopening and implies that the C-terminal ends of the S4 and S6 domains are situated in close proximity at negative transmembrane potentials. However, our data do not provide evidence for or against the possibility that charge pairing between Asp⁵⁴⁰ and Arg⁶⁶⁵ stabilizes the closed state of WT HERG channels.

To further explore the potential importance of electrostatic repulsion between specific residues in the S4-S5 linker and S6 we tested whether introduction of an acidic residue at position 665 would repel Asp⁵⁴⁰ and allow channels to reopen with hyperpolarization. We were unable to reproduce the property of hyperpolarization-induced channel gating by engineering acidic residues at positions 540 and 665. A possible explanation for this negative result relates to the relative length of the side chains of the residues at these positions. The side chains of Arg and Lys are substantially longer than Asp. Perhaps the side chains of Asp⁵⁴⁰ and Asp⁶⁶⁵ are not close enough or favorably oriented for electrostatic repulsion to occur and cause channel reopening.

Although R665D HERG channels did not reopen with hyperpolarization, we did observe a dramatic effect on channel gating. Deactivation was markedly slower for R665D HERG compared with WT HERG channels, far in excess than what could be explained by the shift in voltage dependence of activation. Thus, once the open conformation is achieved by depolarization, the transitions to the closed conformation are inhibited by Asp⁶⁶⁵. Other amino acid substitutions at position 665 (R665A, R665Q) resulted in qualitatively similar but less dramatic effects. In a previous study, we individually mutated residues 646-667 to Ala in the S6 domain of HERG (21). Channel deactivation was not significantly affected with the exception of V659A and R665A. V659C channels were also reported to deactivate slowly (22). It is possible that amino acid substitutions at specific residues in the S6 domain of HERG slow channel deactivation by impeding rotation of the helical bundle back into its resting conformation.

It is conceivable that mutations at positions 540 and 665 may exert independent effects on channel gating. For example, mutation of Asp⁵⁴⁰ to Lys or Arg could enable inward movement of the S4 domain and cause channel reopening without interacting specifically with Arg⁶⁶⁵. An additional mutation of Arg⁶⁶⁵ to a nonbasic residue might independently stabilize the closed configuration and prevent the D540K-induced channel reopening. Another possible explanation for our results is that Lys⁵⁴⁰ and Arg⁶⁶⁵ may interact by an allosteric mechanism.

Coupling between Voltage Sensing and Channel Activation-- How is membrane hyperpolarization coupled to channel opening for D540K HERG but not WT HERG? One possibility is that the S4 domain in WT HERG moves inward with hyperpolarization but does not couple to channel opening at negative potentials unless Asp⁵⁴⁰ is mutated to Lys. This seems unlikely because inward gating currents are not measurable below 120 mV in other Kv channels, including ether-a-go-go channels that share high homology with HERG (5, 6, 23). More likely, inward movement of the S4 domain is enabled by the D540K mutation.

The time and voltage dependence of hyperpolarization-induced activation of D540K HERG is qualitatively similar to activation of HCN pacemaker channels. This suggested to us that an interaction between the S4-S5 linker and the S6 domain could mediate activation of HCN channels. Indeed, several mutations in the S4-S5 linker of HCN channels disrupted normal channel closure as though the link between voltage-sensing and channel opening was uncoupled (24). Thus, the S4-S5 linker may physically couple S4 movement with the opening of both HERG and HCN channels.

Further support for the importance of interactions between the S4-S5 linker and the C-terminal portion of S6 in channel gating comes from studies of chimeric Shaker and KcsA channels. Chimeric channels containing the pore module of KcsA (inner helix, pore domain, outer helix, and C terminus) inserted into the background of the voltage-sensing module of Shaker (S1-S4 and S4-S5 linker) traffic normally but do not gate in response to membrane depolarization (25). However, if the C terminus of Shaker is included in the chimera, the resulting channel opens in a voltage-dependent fashion (26). The specific residues of the C terminus that perform a critical role in transduction of voltage sensing to chimeric channel activation remain to be determined but may be analogous to the interactions described here for D540K HERG.

In summary, the hyperpolarization-dependent reopening of D540K HERG channels is mediated by charge repulsion between Lys⁵⁴⁰ located in the S4-S5 linker and Arg⁶⁶⁵ located in the C-terminal end of the S6 domain. The bell-shaped voltage dependence of activation suggests that the S4 domain of D540K HERG can move inward in response to membrane hyperpolarization. We speculate that inward movement of the S4 domain with hyperpolarization is normally restricted by a stabilizing interaction between the S4-S5 linker and the C-terminal region of S6. Finally, our findings that Lys⁵⁴⁰ interacts with Arg⁶⁶⁵ puts constraints on the possible conformation of the HERG channel with respect to the location of the C-terminal ends of the S4 and S6 domains.

	ACKNOWLEDGEMENTS

We thank Peter Westenskow and Monica Lin for valuable technical support.

	FOOTNOTES

^* This work was supported by National Heart, Lung, and Blood Institute Grants HL03816 (to M. T.-F.) and HL65299 (to M. C. S.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Pediatric Cardiology, Suite 1500 PCMC, University of Utah School of Medicine, 100 N. Medical Dr., Salt Lake City, UT 84113. Tel.: 801-588-2604; Fax: 801-588-2612; E-mail: mfirouzi@hmbg.utah.edu.

Published, JBC Papers in Press, February 25, 2002, DOI 10.1074/jbc.M200410200

	ABBREVIATIONS

The abbreviations used are: HERG, human ether-a-go-go-related gene; Kv, voltage-gated potassium; O, open state; O_h, hyperpolarization-induced open state; WT, wild-type; MES, 2-(N-morpholino)ethanesulfonic acid; HCN, hyperpolarization-activated channel.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. J. Labro, A. L. Raes, A. Grottesi, D. Van Hoorick, M. S.P. Sansom, and D. J. Snyders Kv Channel Gating Requires a Compatible S4-S5 Linker and Bottom Part of S6, Constrained by Non-interacting Residues J. Gen. Physiol., December 1, 2008; 132(6): 667 - 680. [Abstract] [Full Text] [PDF]

				S. L. Wynia-Smith, A. L. Gillian-Daniel, K. A. Satyshur, and G. A. Robertson hERG Gating Microdomains Defined by S6 Mutagenesis and Molecular Modeling J. Gen. Physiol., October 27, 2008; 132(5): 507 - 520. [Abstract] [Full Text] [PDF]

				H. Sale, J. Wang, T. J. O'Hara, D. J. Tester, P. Phartiyal, J.-Q. He, Y. Rudy, M. J. Ackerman, and G. A. Robertson Physiological Properties of hERG 1a/1b Heteromeric Currents and a hERG 1b-Specific Mutation Associated With Long-QT Syndrome Circ. Res., September 26, 2008; 103(7): e81 - e95. [Abstract] [Full Text] [PDF]

				C. Alonso-Ron, P. de la Pena, P. Miranda, P. Dominguez, and F. Barros Thermodynamic and Kinetic Properties of Amino-Terminal and S4-S5 Loop HERG Channel Mutants under Steady-State Conditions Biophys. J., May 15, 2008; 94(10): 3893 - 3911. [Abstract] [Full Text] [PDF]

				I. R. Boulet, A. J. Labro, A. L. Raes, and D. J. Snyders Role of the S6 C-terminus in KCNQ1 channel gating J. Physiol., December 1, 2007; 585(2): 325 - 337. [Abstract] [Full Text] [PDF]

				R. M. Hardman, P. J. Stansfeld, S. Dalibalta, M. J. Sutcliffe, and J. S. Mitcheson Activation Gating of hERG Potassium Channels: S6 GLYCINES ARE NOT REQUIRED AS GATING HINGES J. Biol. Chem., November 2, 2007; 282(44): 31972 - 31981. [Abstract] [Full Text] [PDF]

				M. Perry, F. B. Sachse, and M. C. Sanguinetti Structural basis of action for a human ether-a-go-go-related gene 1 potassium channel activator PNAS, August 21, 2007; 104(34): 13827 - 13832. [Abstract] [Full Text] [PDF]

				Q. Ouyang, M. Goeritz, and R. M. Harris-Warrick Panulirus interruptus Ih-Channel Gene PIIH: Modification of Channel Properties by Alternative Splicing and Role in Rhythmic Activity J Neurophysiol, June 1, 2007; 97(6): 3880 - 3892. [Abstract] [Full Text] [PDF]

				J. B. Saenen, A. J. Labro, A. Raes, and D. J. Snyders Modulation of HERG Gating by a Charge Cluster in the N-Terminal Proximal Domain Biophys. J., December 15, 2006; 91(12): 4381 - 4391. [Abstract] [Full Text] [PDF]

				A. Lampert, S. D. Dib-Hajj, L. Tyrrell, and S. G. Waxman Size Matters: Erythromelalgia Mutation S241T in Nav1.7 Alters Channel Gating J. Biol. Chem., November 24, 2006; 281(47): 36029 - 36035. [Abstract] [Full Text] [PDF]

				A. Bhattacharji, B. Kaplan, T. Harris, X. Qu, M. W. Germann, and M. Covarrubias The Concerted Contribution of the S4-S5 Linker and the S6 Segment to the Modulation of a Kv Channel by 1-Alkanols Mol. Pharmacol., November 1, 2006; 70(5): 1542 - 1554. [Abstract] [Full Text] [PDF]

				D. L. Prole and G. Yellen Reversal of HCN Channel Voltage Dependence via Bridging of the S4-S5 Linker and Post-S6 J. Gen. Physiol., August 28, 2006; 128(3): 273 - 282. [Abstract] [Full Text] [PDF]