Molecular mapping of a site for Cd2+-induced modification of human ether-à-go-go-related gene(hERG) channel activation
1 Department of Physiology and Nora Eccles Harrison Cardiovascular Research and Training Institute, University of Utah, Salt Lake City, UT 84112-5000, USA
Abstract
Cd2+ slows the rate of activation, accelerates the rate of deactivation and shifts the half-points of voltage-dependent activation (V0.5,act) and inactivation (V0.5,inact) of human ether-à-go-go-related gene (hERG) K+ channels. To identify specific Cd2+-binding sites on the hERG channel, we mutated potential Cd2+-coordination residues located in the transmembrane domains or extracellular loops linking these domains, including five Cys, three His, nine Asp and eight Glu residues. Each residue was individually substituted with Ala and the resulting mutant channels heterologously expressed in Xenopus oocytes and their biophysical properties determined with standard two-microelectrode voltage-clamp technique. Cd2+ at 0.5 mM caused a +36 mV shift of V0.5,act and a +18 mV shift of V0.5,inact in wild-type channels. Most mutant channels had a similar sensitivity to 0.5 mM Cd2+. Mutation of single Asp residues located in the S2 (D456, D460) or S3 (D509) domains reduced the Cd2+-induced shift in V0.5,act, but not V0.5,inact. Combined mutations of two or three of these key Asp residues nearly eliminated the shift induced by 0.5 mM Cd2+. Mutation of D456, D460 and D509 also reduced the comparatively low-affinity effects of Ca2+ and Mg2+ on V0.5,act. Extracellular Cd2+ modulates hERG channel activation by binding to a coordination site formed, at least in part, by three Asp residues.
, http://www.100md.com
Introduction
Human ether-à-go-go-related gene (hERG) channel is a member of the ether-à-go-go (EAG) family of voltage-gated K+ channels (Warmke & Ganetzky, 1994) and is expressed in many cell types, including cardiac myocytes, neurones and tumour cells. In cardiomyocytes, hERG channel subunits co-assemble to form homotetrameric channels that conduct the rapid delayed rectifier K+ current IKr (Sanguinetti et al. 1995; Trudeau et al. 1995). IKr is an important component of the outward currents that mediate repolarization of the cardiac action potential. Inherited loss-of-function mutations in HERG or block of hERG channels by specific drugs prolongs the duration of action potentials and causes long QT syndrome, a disorder of ventricular repolarization that increases the risk of life-threatening arrhythmias (Keating & Sanguinetti, 2001).
, 百拇医药
Divalent cations can modify ion channel gating by screening non-specific negative surface charges located on membrane lipids and proteins (McLaughlin, 1989) or by binding to specific residues of channel proteins (Gilly & Armstrong, 1982). Recent studies have attempted to localize the binding sites using site-directed mutagenesis. Several acidic residues in the S5–P linker of Kv1–Kv4 channels were proposed to account for the majority of negative surface charges that were screened by divalent cations (Elinder et al. 1996, 1998). In Drosophila EAG (dEAG) channels, Mg2+, Mn2+ and Ni2+ slow gating by binding to an extracellular site formed by two acidic residues: D278 in S2 and D327 in S3 (Silverman et al. 2000). Charge pair interactions between these Asp residues and specific basic residues in S4 have been proposed to facilitate activation of dEAG channels and protein folding and activation of Shaker channels (Papazian et al. 1995; Seoh et al. 1996; Tiwari-Woodruff et al. 1997, 2000; Silverman et al. 2000, 2003, 2004). Specific charge pairings may be disrupted and replaced by others as the S4 helix slides past the relatively immobile S2 and S3 domains. In this scheme, divalent cations bind to the Asp residues in S2 or S3 and prevent interaction with S4, causing a slowing of EAG channel activation.
, http://www.100md.com
Only a few of the many residues in hERG channel that could potentially interact with divalent cations have been investigated and there is no consensus regarding the location of a binding site. Asp residues in S2 and S3 of dEAG are conserved in hERG. D278, D282 and D327 of dEAG are homologous to D456, D460 and D509 of hERG channel (Fig. 1A)
A, sequence alignment of S2 and S3 domains of hERG and EAG channels. B, topology of a single hERG channel subunit with six transmembrane domains (S1–S6) that span the membrane. The approximate location of the Asp, Glu, His and Cys residues mutated to Ala in this study are shown. The residues that interact with Cd2+ are highlighted in large italic text.
, 百拇医药
The conservation of charged residues in S2 and S3 suggests that divalent cations could bind to a common site on dEAG and hERG channels and affect gating in a similar manner. However, several recent studies argue against these simple predictions. First, Mg2+ greatly slows the rate of dEAG channel opening, but unlike in hERG, does not cause a shift in the voltage dependence of activation (Tang et al. 2000). Second, mutation of the acidic residues of S2 and S3 to Cys did not alter Ca2+-induced changes in hERG channel gating (Liu et al. 2003). Third, the shift in hERG gating by Ca2+ is influenced by an acidic residue (E519) located in the extracellular S3–S4 linker and neutralization of an adjacent residue (E518) shifts the voltage dependence of activation to more positive potentials and increases the affinity of the channel for Ca2+ (Johnson et al. 2001). Fourth, more than one binding site may account for the multiple effects of divalent cations on hERG or dEAG channel gating. For example, mutation of D327 to Ala in dEAG eliminates Mg2+ but not Ni2+ sensitivity, suggesting that the coordination sites for Ca2+ and Ni2+ overlap, but are not identical (Silverman et al. 2004). The structural basis of divalent cation-mediated changes in hERG channel gating is unclear and a more systematic study is warranted.
, http://www.100md.com
Alkaline earth cations (e.g. Ca2+ and Mg2+) slow the rate of activation, accelerate the rate of deactivation and induce positive shifts in the voltage dependence of hERG channel activation, and to a lesser extent, inactivation (Ho et al. 1996, 1998; Johnson et al. 1999, 2001; Silverman et al. 2000, 2003). Transition metal ions (e.g. Cd2+, Co2+, Ni2+ and Zn2+) have similar effects on hERG channel gating, but are more potent (Anumonwo et al. 1999; Johnson et al. 1999; Sanchez-Chapula & Sanguinetti, 2000). Previous reports have suggested that divalent cations may differentially affect hERG channel activation and inactivation. For example, low concentrations of Cd2+ (< 200 μM) preferentially shifted the voltage dependence of hERG channel inactivation, causing reduced rectification of the current–voltage (I–V) relationship and an increased peak outward current, whereas higher Cd2+ concentrations were required to affect activation (Johnson et al. 1999).
, http://www.100md.com
The coordination geometry of transition metal ions can be complex. Based on a comprehensive survey of binding residues in metalloproteins (Rulisek & Vondrasek, 1998), the typical coordination geometry of Cd2+ is either tetrahedral or octahedral and can involve up to five different amino acid residues with the following relative order of importance: Cys > Asp > His > Glu > Asn. Most often, one or two sites in the tetrahedral, and more than two sites in the octahedral, coordination complex are water molecules. To identify Cd2+-binding sites on hERG channel, it is important to assay all potential binding residues because the 3-dimensional structure of the channel is incompletely understood. Towards this goal, we have mutated all 25 amino acids that could conceivably form a coordination site for extracellular Cd2+, including nine Asp, eight Glu, three His and five Cys residues (Fig. 1B). Residues were mutated to Ala and the resulting mutant channels expressed in Xenopus oocytes. The voltage dependence and kinetics of activation and inactivation of mutant channels were compared to wild-type (WT) hERG channels using two-microelectrode voltage-clamp technique. Mutation of three Asp residues in S2 and S3 caused a positive shift in V0.5,act and selectively prevented further shifts normally induced by extracellular Cd2+. Mutation of three amino acids located in the S5–pore linker only affected the Cd2+-induced shift in V0.5,inact. Together these findings suggest that Cd2+ binds to different sites on hERG channel to differentially affect channel activation and inactivation.
, http://www.100md.com
Methods
Molecular biology
hERG channel was subcloned into the pSP64 plasmid expression vector (Promega, Madison, WI, USA) and site-directed mutagenesis was performed as previously described (Mitcheson et al. 2000). Mutations were confirmed by restriction mapping and DNA sequencing. Complementary RNA (cRNA) for injection into oocytes was prepared with SP6 Cap-Scribe (Roche Applied Science, Indianapolis, IN, USA) following linearization of the expression construct with EcoRI.
, http://www.100md.com
Oocytes and solutions
Oocytes were isolated by dissection from anaesthetized adult Xenopus laevis. Frogs were anaesthetized by immersion in 0.2% tricaine (Sigma) for 10–15 min. A small abdominal incision was made and ovarian lobes containing oocytes were removed. The incision was sutured closed and the frog returned to its aquarium for a recovery period of at least 1 month before the procedure was repeated. After a maximum of three surgical procedures, tricaine-anaesthetized frogs were killed by pithing. Clusters of oocytes were treated with 2 mg l–1 type 2 collagenase (Worthington) to remove follicle cells. Maintenance and cRNA injections into oocytes were performed as previously described (Stuhmer, 1992; Sanguinetti & Xu, 1999). During voltage-clamp experiments, the oocytes were bathed in a solution containing, (mM): NaCl 96, KCl 2, CaCl2 1, MgCl2 2 and Hepes 5; pH adjusted to 7.6 with NaOH. Stock solutions of CdCl2 (Sigma) were prepared in this solution at concentrations of 0.1 or 1 M and kept at 4°C. CdCl2 remained soluble in the saline solution at 30 mM, the highest concentration used in this study. No compensation was made for changes in extracellular osmolarity that resulted from addition of CdCl2 to the bathing solution or for Cd2+ binding to Hepes (Cherny & DeCoursey, 1999). The solution flow rate was 1.5 ml min–1 and the oocyte chamber volume was 0.2 ml.
, 百拇医药
Voltage-clamp and data analysis
Ionic currents were recorded at room temperature (22–24°C) using a GeneClamp 500 amplifier (Molecular Devices, Sunnyvale, CA, USA) and standard two-microelectrode voltage-clamp technique (Stuhmer, 1992). Isochronal I–V relationships were determined with 2-s test pulses applied at a frequency of 0.06 Hz and in 10-mV increments to voltages ranging from –70 mV up to +120 mV. The holding potential was –80 mV. The amplitude of current measured at the end of the 2-s pulse was normalized to the peak value measured under control conditions (before exposure of an oocyte to CdCl2).
, http://www.100md.com
Fully activated I–V relationships were determined from analysis of tail currents. A 2-s prepulse to a voltage where activation was complete for each mutant channel and condition (indicated in figure legends) was applied before repolarization to a variable test potential. Tail currents during the test potential were fitted to a bi-exponential function using the Chebyshev method for curve fitting provided by Clampfit 8 software (Molecular Devices, Sunnyvale, CA). The peak tail current (Itail) amplitude was determined by extrapolation of fitted current traces back to the moment of membrane repolarization. Itail for each pulse was normalized to the largest value (Itail,max) measured under control conditions and plotted as a function of test voltage (Vt). The kinetics of channel deactivation were quantified by time constants and relative amplitudes of the fast and slow components of tail current decay.
, http://www.100md.com
The conductance–voltage (g–V) relationship for hERG channel current was determined from peak Itail elicited after 2-s pulses to a variable test potential. Itail for each test potential was normalized to Itail,max. The resulting relative conductance (g/gmax) values were plotted against Vt, and the relationship fitted to a Boltzmann function (eqn (1)) using Origin 7 software (OriginLab, Northampton, MA, USA).
where V0.5,act is the voltage required to elicit 0.5 x g/gmax and k is the slope factor of the relationship.
, 百拇医药
The voltage dependence for the recovery from hERG channel inactivation was determined with a three-pulse protocol (Smith et al. 1996; Spector et al. 1996). First, hERG channels were activated or inactivated by a 1-s prepulse to a potential (Vpre) that was varied from +40 to +90 mV depending on the specific mutation. In a second pulse, channels were allowed to recover from inactivation during a 10-ms pulse applied to a test voltage (Vt) that was increased in 10-mV steps from –140 mV to the voltage used for the first pulse. During the third and final pulse, channels were inactivated by stepping the membrane potential back to Vpre and the resulting ionic current following the decay of the capacitance current was fitted with a mono-exponential function using the Levenberg-Marquardt method in pCLAMP software (Molecular Devices). The amplitude of inactivating current (Iinact) was determined by extrapolation of the fitted current trace back to the initial time of the third pulse. Iinact was normalized to the largest value (Iinact,max) and plotted as a function of Vt. Iinact – Vt curves were fitted with a Boltzmann function similar to eqn (1) to estimate V0.5,inact and k for the relationship. In this study, we use the term ‘inactivation’ to refer to the voltage dependence of recovery from inactivation.
, 百拇医药
The initial screening of mutant channels utilized an extracellular Cd2+ concentration ([Cd2+]e) of 0.5 mM. Some of the mutant channels were further characterized by determining the effect of 10 μM to 30 mM [Cd2+]e. Concentration–effect data were fitted to the Hill equation (eqn (2)) to determine the EC50, the effective [Cd2+]e required for the shift in V0.5,act (Vact) to reach half its maximum value (Vmax), and the Hill coefficient (h).
Data in tables are presented as mean ± S.E.M. (n = number of cells) and statistical comparisons between experimental groups were performed using the Student's t test. Differences were considered significant at P < 0.05.
, 百拇医药
Results
Effects of Cd2+ on wild-type hERG channels
We first determined the effects of a single concentration of Cd2+ (0.5 mM) on hERG channels expressed in Xenopus oocytes. Extracellular application of 0.5 mM Cd2+ caused an immediate change in the kinetics and amplitude of WT hERG channel currents elicited by a 2-s pulse to +60 mV from a holding potential of –80 mV (Fig. 2A). The most notable effects were a slowing of activation (Fig. 2B) and accelerated deactivation at –70 mV (Fig. 2C). These effects were rapidly and completely reversible upon washout of Cd2+ from the bathing solution (not shown), indicating an extracellular site of action. Currents elicited in response to 2-s test potentials ranging from –70 to +70 mV are shown in Fig. 2D. Currents measured at the end of each pulse were normalized relative to the peak values at –10 mV under control conditions and the averages for 11 oocytes plotted as a function of test potential in Fig. 2E. Cd2+ increased current magnitudes for test potentials positive to +20 mV, but decreased currents at test potentials negative to +20 mV. Cd2+ also shifted the peak of the I–V relationship by 35 mV, equivalent to the positive shift in the voltage dependence of activation (V0.5,act, +36.4 ± 0.8, n = 11) determined by plotting the normalized tail current amplitude as a function of test potential (Fig. 2F).
, 百拇医药
A, currents recorded during 2-s pulses to +60 mV before (thin trace) and in the presence (thick trace) of 0.5 mM Cd2+. Tail currents were measured at –70 mV. B and C, the currents inside the dashed boxes in A are shown in an expanded view. D, currents recorded at test potentials of –50 to +50 mV, applied in 20-mV increments before (left) and after (right) exposure to Cd2+. E, normalized isochronal I–V relationships for currents measured at end of 2-s pulses before () and after application of Cd2+ () (n = 11). F, activation curves determined from peak tail currents following 2-s pulses to the indicated potentials. G, currents recorded before (left) and after (right) using a two-pulse voltage-clamp protocol shown in inset. To ensure currents were fully activated, the first pulse was to +30 mV in the absence of Cd2+ (left) and +60 mV after adding 0.5 mM Cd2+ (right). Current traces correspond to test potentials of –120 to +30 mV applied in 30-mV increments. H, fully activated I–V relationships for peak (extrapolated) tail currents using protocol shown in G. I, relative amplitude of fast component of deactivating current. J and K, time constants for current deactivation. Symbols in F and H–K are the same as in E.
, 百拇医药
The kinetics of hERG channel current deactivation were determined over a range of membrane potentials. The oocyte was first depolarized to +60 mV for 2 s before repolarization to a potential that was varied from +30 to –140 mV. Examples of currents recorded with this pulse regimen before, and 3 min after, addition of 0.5 mM Cd2+ to the extracellular solution are shown in Fig. 2G. The peak currents during the second pulse were plotted as a function of voltage to obtain the ‘fully activated’ I–V relationship (Fig. 2H). These I–V relationships exhibit marked inward rectification because channels only partially recover from rapid C-type inactivation at potentials positive to –100 mV. Tail currents were fitted with a bi-exponential function to determine the relative amplitudes (Fig. 2I), and the time constants of the fast and slow components of current decay (Fig. 2J and K). Cd2+ accelerated the rate of deactivation at all potentials examined and caused more than a +30-mV shift in the voltage dependence of these measures of deactivation, similar to the Cd2+-induced shift in V0.5,act. The increase in outward currents by Cd2+ at potentials positive to –60 mV in the fully activated I–V relationship (Fig. 2H) suggested an effect on inactivation gating. Therefore, the effects of 0.5 mM Cd2+ on the voltage dependence of recovery from C-type inactivation were determined with a three-pulse protocol shown in Fig. 3A (Smith et al. 1996; Spector et al. 1996). Cd2+ shifted the half-point for this relationship (V0.5,inact) by +18 mV (Fig. 3B), a shift that was half the Cd2+-induced effect on channel activation. The effect of Cd2+ on activation gating caused a shift in the peak of the I–V relationship from –10 to +20 mV. The lesser effect of Cd2+ on inactivation gating reduced the peak of the I–V relationship with a cross-over at potentials > +20 mV (Fig. 2E).
, http://www.100md.com
A, representative current recordings of WT hERG channel before (left) and after (right) exposure to 0.5 mM Cd2+. Currents shown were elicited by pulsing to test potentials from –130 to –30 mV applied in 20-mV increments after a 1-s prepulse to +40 mV before and +60 mV in the presence of Cd2+ (not shown). Inset shows the voltage changes during the second of the three pulses. B, plot of fractional availability of channels to open (relative amplitude of extrapolated currents elicited during third step of the three-pulse protocol) in the absence () and presence of 0.5 mM Cd2+ (). The V0.5,inact was –89.4 mV and k was 22.1 mV for control and –71.4 mV and 22.4 mV, respectively, after exposure of oocytes to 0.5 mM Cd2+. C, Cd2+ binds to and modifies gating of inactivated hERG channels. Currents were recorded during 40-s pulses to +50 mV before (trace i), during (trace ii) and after (trace iii) exposure of the oocyte to 0.5 mM Cd2+. The solution switch was made at the time indicated by the arrow during trace ii. The time between pulses was 15 s.
, http://www.100md.com
The rate of hERG channel activation was slower during the first depolarizing pulse applied immediately after switching to a bathing solution containing Cd2+. This finding suggests that Cd2+ can bind to the closed state of the channel to affect gating associated with a subsequent voltage-dependent activation. To determine whether modification of gating can occur when the channel is in other states, we switched the control bathing solution to one containing 0.5 mM Cd2+ during a prolonged (40 s) voltage-clamp pulse to +50 mV, a voltage where all channels were predicted to be in either the open or inactivated state. In the example shown in Fig. 3C, the solution switch was made 10 s after the onset of the 40-s pulse. The current amplitude increased immediately (Fig. 3C, trace ii), then slowly enhanced over the next 30 s to a new steady-state level. The increase in current magnitude at +50 mV was consistent with some channels recovering from an inactivated state to an open state and suggests that Cd2+ destabilized the inactivated channel state.
, 百拇医药
Effects of Cd2+ on inactivation-deficient hERG channels
The double mutation G628C/S631C eliminates fast inactivation of the hERG channel (Smith et al. 1996). At low concentrations, Cd2+ was reported to have only minor and insignificant effects on current amplitude or kinetics of deactivation of this mutant channel, suggesting that channel inactivation might be essential for a Cd2+-induced increase in hERG channel current (Johnson et al. 1999). We determined the effects of 0.5 mM Cd2+ on G628C/S631C hERG channels. Cd2+ increased current magnitude at potentials positive to +20 mV (Fig. 4A and B). The mechanism of this effect is unknown, but it is possible that Cd2+ might bind to these introduced Cys residues and affect pore properties (e.g. increased single channel conductance or mean open time). Cd2+ (0.5 mM) shifted V0.5,act by +38 mV (Fig. 4C), the same shift as that observed with WT hERG channels. Cd2+ also slowed the rate of current activation (Fig. 4A and D) and accelerated the rate of current deactivation (Fig. 4E) at all potentials examined. For example, the fast and slow time constants (fast and slow, respectively) for deactivation at –110 mV were 51 ± 8 ms and 314 ± 7 ms in control, compared to 16 ± 3 and 130 ± 16 ms after Cd2+ (n = 3), respectively. At –80 mV, fast was 212 ± 58 ms and slow was 656 ± 166 ms in control, compared to a fast value of 38 ± 6 ms and a slow value of 163 ± 14 ms after application of Cd2+ (n = 3). Thus, Cd2+-induced changes in deactivation and shifts in the voltage dependence of activation gating of inactivation-deficient G628C/S631C channels were the same as WT hERG channels.
, 百拇医药
A, currents conducted by G628C/S631C channels before (left) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (right). Currents were elicited with 2-s pulses to potentials from –50 mV to +50 mV, applied in 20-mV increments. Tail currents were measured at –120 mV. B, normalized I–V relationships for peak currents elicited with 2-s pulses before () and after application of 0.5 mM Cd2+ (). Currents were normalized relative to the peak current at +50 mV in the absence of Cd2+. C, g–V relationship before () and after application of 0.5 mM Cd2+ (). Control: V0.5,act, –27.5 ± 0.6 mV; k, 11.0 ± 0.3 mV; after application of Cd2+: V0.5,act, +10.4 ± 0.6 mV; k, 13.9 ± 2.5 mV (n = 4). D, time constants for activation of G628C/S631C hERG channel currents before (open symbols) and after application of Cd2+ (filled symbols). E, tail currents recorded at potentials of –120 to –30 mV, applied in 10-mV increments. Top, voltage pulse protocol; middle, control currents; bottom, currents in presence of 0.5 mM Cd2+.
, 百拇医药
Site-directed mutagenesis of potential Cd2+-binding residues in the extracellular linkers of the hERG channel
The coordination geometry of Cd2+ in metalloproteins is either tetrahedral or octahedral and the most frequent residues involved in Cd2+ coordination are Cys, His, Asp and Glu (Rulisek & Vondrasek, 1998). Such residues that were accessible from the extracellular solution and could potentially interact with extracellular Cd2+ were mutated to Ala. Residues located on the N-terminus, C-terminus or intracellular linkers were assumed to be inaccessible to extracellular Cd2+ during the short timecourse of the experiments and therefore, were not investigated. The approximate locations of the Ala-substituted residues, including five Cys, three His, nine Asp and eight Glu are illustrated in Fig. 1B. Residues that when mutated to Ala reduced the 0.5 mM Cd2+-induced shift in the voltage dependence of hERG channel activation were considered to be potential coordination sites for Cd2+. A Cd2+ concentration of 0.5 mM was used because it was high enough to cause a large shift in V0.5,act (+36 mV) of WT channels, yet low enough to minimize non-specific binding and screening of negative surface charges. Mutation of a critical Cd2+-binding residue would be expected to reduce binding affinity. However, because 0.5 mM is near the EC50 for the effects of Cd2+ on WT hERG channels, a mutation-induced change in the shift in V0.5,act could reflect a change in binding affinity, maximum shift or a combination of the two effects.
, http://www.100md.com
Fourteen candidate Cd2+-binding residues are located in the extracellular linkers of hERG channel (Fig. 1B). Single or combined mutations were used to assess the potential role of these residues in Cd2+ binding. Four Glu and two Cys residues are located in the S1–S2 linker. Instead of assaying six different channels containing single mutations, we constructed a mutant channel with all six residues mutated to Ala (‘4E2C/6A’ hERG). This mutant channel conducted currents that were similar to WT hERG channels. It is surprising that neutralization of 16 negative charges (4 Glu x 4 subunits) near the extracellular surface of each channel did not alter V0.5,act or the response to 0.5 mM Cd2+. The Cd2+-induced shifts in the isochronal (Fig. 5B) and fully activated I–V relationships (Fig. 5C), V0.5,act (Table 1) and deactivation kinetics (Fig. 5D) of 4E2C/6A hERG were the same as in WT hERG channels. The 4E2C/6A mutations did not alter the Cd2+-induced shift in V0.5,inact (Table 2). Thus, mutation of all the residues in the S1–S2 linker to Ala did not alter the effects of Cd2+ on channel gating.
, 百拇医药
A, currents conducted by 4E2C/6A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from –70 mV to +30 mV, applied in 20-mV increments. B–D, effect of 0.5 mM Cd2+ on 4E2C/6A currents: isochronal I–V relationships normalized to peak current (Ipeak) under control conditions (B); fully activated I–V relationships normalized to the peak tail current (Itail) measured under control conditions (C); and time constants of current deactivation for 4E2C/6A currents before and after application of Cd2+. Left-hand axis indicates values for slow and right-hand axis is for fast (D). In each panel, open symbols represent data for control currents and filled symbols represent measurements in the presence of 0.5 mM Cd2+. E–H, the same protocols and analyses for E518A hERG channel currents. Current traces in E correspond to test potentials of –60 to +60 mV in 20-mV increments for both conditions, before (top) and after adding 0.5 mM Cd2+ (bottom). I–P, the same protocols and analyses for D580A (I–L) and H587A (M–P) hERG channel currents. Current traces for D580A (I) and H587A (M) mutant channels correspond to test potentials of –50 to +50 mV applied in 20-mV increments. Note the lack of effect of 0.5 mM Cd2+ on fully activated I–V relationship for D580A mutant channels (K) that results from a reduced effect on inactivation (Table 2).
, 百拇医药
Two candidate Cd2+-binding residues (E518 and E519) are located in the S3–S4 linker. Mutation of these residues were previously reported to alter the extracellular Ca2+-induced changes in hERG channel gating (Johnson et al. 2001). We found that the biophysical properties of E519A hERG channels were similar to WT, whereas the E518A mutation shifted the peak of the I–V relationship (Fig. 5F) and V0.5,act by +26 mV (Table 1) and accelerated the rate of deactivation (Fig. 5H). Despite differences in biophysical properties, E518A (Fig. 5E–H) and E519A (not shown) mutant channels retained a normal 0.5 mM Cd2+-induced effect on activation (Table 1) and inactivation (Table 2).
, 百拇医药
There are six potential Cd2+-interacting residues located in the S5–P region, including E575, H578, D580, H587, D591 and D609 (Fig. 1B). A previous study reported that mutation of some of these residues to Cys (i.e. E575, H578, D580 and D609) did not affect inactivation, whereas D591C and H587C disrupted inactivation (Liu et al. 2002). Mutation of these residues to Ala did not significantly alter the baseline biophysical properties of the channels, or the effect of 0.5 mM Cd2+ on V0.5,act compared to WT the hERG channel (Table 1). For example, current traces, I–V relationships and kinetics of deactivation are shown in Fig. 5I–L for D580A channels and in Fig. 5M–P for H587A channels. The effects of Cd2+ on H587A channel properties were normal. However, Cd2+ reduced D580A tail-current amplitudes by 16% (Fig. 5I) and did not cause a positive shift in the voltage dependence of the fully activated I–V relationship (Fig. 5K). E575A and H578A channel currents were also reduced by Cd2+. The V0.5,inact for E575A, H578A and D580A was only shifted by +7 to +11 mV in response to 0.5 mM Cd2+ (Table 2). Combining two of the mutations (H578A/D580A) or all three mutations (E575/H578/D580) into a single channel did not have an additive effect. The Cd2+-induced shift in V0.5,inact for the double and triple mutants was +8 mV, similar to the shift caused by single mutations (Table 2).
, 百拇医药
In summary, mutations of the 14 residues located in the extracellular linkers of hERG channel did not significantly alter Vact induced by Cd2+. However, mutation of E575, H578 or D580 reduced the Cd2+-induced shift in V0.5,inact, confirming the important role of the S5-P linker in rapid C-type inactivation of hERG channels (Liu et al. 2002).
Site-directed mutagenesis of potential Cd2+-binding residues in the transmembrane domains of the hERG channel
, 百拇医药
The only potential Cd2+-interacting residue located in S1 is D411 (Fig. 1B). Mutation of this residue to Ala caused a –17.5 mV shift in V0.5,act (Table 1), a +15 mV shift in V0.5,inact (Table 2), an enhanced rate of activation and deactivation (Fig. 6A and D), and a slight leftward displacement of the isochronal and fully activated I–V relationship (Fig. 6B and C). The Vact for D411A channels was identical to the value for WT channels (+36 mV), but the Vinact was 10 mV greater for the mutant channel (+28 mV) than for the WT channel (+18 mV). There are three potential Cd2+-interacting residues located in S5: C555, H562 and C566 (Fig. 1B). No hERG channel currents were detected in oocytes expressing H562A channels. The biophysical properties of C566A channel currents and their response to 0.5 mM Cd2+ were similar to WT channels (Fig. 6E–H and Tables 1 and 2). C555A (not shown) was also similar to WT. Two potential Cd2+-interacting residues (E637 and C643) are located in S6 (Fig. 1B). E637A channels did not functionally express, but C643A channels had normal biophysical properties and responsiveness to Cd2+ (Fig. 6I–L and Tables 1 and 2). Thus, with the exception of H562A and E637A, which did not functionally express, mutation of Asp or Cys residues located in the S1, S5 and S6 did not diminish the changes in gating caused by Cd2+.
, 百拇医药
A, currents conducted by D411A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from –60 to +40 mV, applied in 20-mV increments. B–D, effect of 0.5 mM Cd2+ on D411A currents: isochronal I–V relationships normalized to peak current under control conditions (B); fully activated I–V relationships normalized to the peak current measured under control conditions (C); and time constants for fast (triangles) and slow (inverted triangles) components of current deactivation for D411A currents. Left-hand axis indicates values for slow and right-hand axis is for fast (D). In each panel, open symbols represent data for control currents and filled symbols represent measurements in the presence of 0.5 mM Cd2+. E–H, the same protocols and analyses for C566A hERG channel currents. Current traces in E correspond to test potentials ranging from –50 to +50 mV applied in 20-mV increments. I–L, the same protocols and analyses for C643A hERG channel currents. Test potentials used for C643A mutant channel (I) are the same as those used for D411A mutant described in A.
, 百拇医药
There are three potential Cd2+-interacting residues located in S2, including D456, D460 and D466 (Fig. 1B). As expected from previous studies of divalent cation interactions with Shaker (Papazian et al. 1995; Tiwari-Woodruff et al. 2000) and dEAG channels (Silverman et al. 2000, 2004), mutation of these Asp residues in hERG channel caused significant changes in channel gating. D456A and D460A channels deactivated more rapidly than normal and the I–V relationships were shifted towards the right compared to the control(Fig. 7A–H). The V0.5,act was shifted by +34 mV for D456A and +22 mV for D460A (Table 1). As in dEAG, mutation of these Asp residues reduced the effects of extracellular Cd2+. The Vact induced by Cd2+ was reduced to +16 mV for D456A and +19 mV for D460A (Table 1). The reduced effect of Cd2+ on Vact together with a normal Vinact (Table 2) caused the peak of the relative I–V relationship in the presence of Cd2+ to be larger compared to other mutant channels (Fig. 7B and F). Moreover, the Cd2+-induced shift in the voltage dependence of deactivation kinetics for these two mutants (Fig. 7D and H) was reduced compared to Vact. Of the three Asp residues in S2, D466 is positioned closest to the C-terminal end of the domain and would be expected to be the least accessible to extracellular divalent cations. Although D466A channels activated and deactivated rapidly (Fig. 7I and L), the V0.5,act was normal (–24 mV) as was the Cd2+-induced shift in the voltage dependence of activation (Vact, +38 mV; Table 1) and deactivation kinetics (Fig. 7L). By contrast, the V0.5,inact of D466A channels was +12 mV more positive than normal, and Cd2+ induced a larger than normal Vinact (+ 28 mV, Table 2). Thus, mutation of the two outermost Asp residues in S2 (D456 and D460) reduced the effects of Cd2+ on hERG channel activation.
, 百拇医药
A, currents conducted by D456A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from –30 to +70 mV, applied in 20-mV increments. B–D, effect of 0.5 mM Cd2+ on D456A currents: isochronal I–V relationships normalized to peak current under control conditions (B); fully activated I–V relationships normalized to the peak current measured under control conditions (C); and time constants for fast (triangles) and slow (inverted triangles) components of current deactivation for D456A currents. Left-hand axis indicates values for slow and right-hand axis is for fast (D). In each panel, open symbols represent data for control currents and filled symbols represent measurements in the presence of 0.5 mM Cd2+. E–H, the same protocols and analyses for D460A hERG channel currents. I–P, the same protocols and analyses for D466A (I–L) and D509A (M–P) hERG channel currents. Examples of current traces for D460A, D466A and D509A mutant channels (E, I and M) correspond to test potentials of –50 to +50 mV applied in 20-mV increments.
, 百拇医药
Two potential Cd2+-binding residues, D501 and D509, are located in the S3 domain (Fig. 1B). D501 channels expressed poorly, but in oocytes with very small endogenous currents we determined that V0.5,act was 42 mV more positive than in WT channels. Despite such a positive V0.5,act, the response to 0.5 mM Cd2+ was equivalent to the response in WT hERG channels, including a +37 mV shift in V0.5,act (Table 1). D509A channels activated at potentials 27 mV more positive than in WT channels and were the least sensitive of all the mutant channels to the effects of Cd2+ on activation gating (Fig. 7M–P). V0.5,act was only shifted +8.1 mV by 0.5 mM Cd2+ (Table 1). By contrast, the V0.5,inact and the Cd2+-induced Vinact values were normal for D509A channels (Table 2). The differential effects of Cd2+ on activation and inactivation gating of D509A channels resulted in an increased peak in the I–V relationship. Mutation of both Asp residues in S3 affected channel gating; however, only mutation of the residue located nearest the extracellular space (D509) diminished the functional response to Cd2+.
, http://www.100md.com
In the majority of metalloproteins, Cd2+ usually binds to Cys residues with equal or greater affinity than to Asp residues (Rulisek & Vondrasek, 1998). Therefore, we individually mutated D456, D460 and D509 to Cys in an attempt to increase the Vact induced by Cd2+. For WT hERG channels, 30 μM and 100 μM Cd2+ induced a Vact of 11.8 ± 0.8 mV (n = 10) and 21.1 ± 0.6 mV (n = 7), respectively. The sensitivity of D460C to Cd2+ was markedly enhanced. For this mutant channel, Vact was increased to 29.3 ± 1.7 mV for 30 μM Cd2+ and 61.3 ± 0.8 mV for 100 μM Cd2+ (n = 5 for each concentration). By contrast, D456C and D509C hERG channels were less responsive than WT channels. The Vact in response to 100 μM Cd2+ was 14.8 ± 1.1 mV for D456C hERG (n = 5) and 13.4 ± 0.5 mV for D509C hERG channels (n = 4). Thus, the sensitivity of hERG channel activation to Cd2+ can be increased by mutation of D460, but not D456 or D509 to Cys.
, 百拇医药
In summary, based on the reduced sensitivity of mutant channels to externally applied Cd2+, two acidic residues in S2 (D456 and D460) and one acidic residue in S3 (D509) contribute to a binding site that mediates the effects of Cd2+ on activation gating of hERG channels. We next determined how mutation of two or all three of these residues affected the response to Cd2+.
Combined mutation of acidic residues in S2 and S3
Three double mutant channels (D456A/D460A, D456A/D509A and D460A/D509A) were constructed. The V0.5,act of D456A/D509A channels was +56 mV more positive than of WT channels and nearly equivalent to the sum of the Vact values induced by the individual mutations (+59 mV). The V0.5,act values for D456A/D460A (+62 mV) and D460A/D509A (+47 mV) channels were shifted rightwards more than expected from the sum of the single mutations. The rate of deactivation was markedly enhanced for all three double mutant channels (Fig. 8A, C and E), similar to the effect caused by mutations of the single residues (Fig. 7). The Vact induced by 0.5 mM Cd2+ was +8.4 mV for D460A/D509A, +4.1 mV for D456A/D509A and +2.5 mV for D456A/D460A channels (Table 1). D460A/D509A channels retained the same sensitivity to Cd2+ as the D509A channels (Vact of +8 mV for both), suggesting that D460 is the least important of the three residues for coordination of Cd2+. At the very positive potentials required to activate the double mutant channels, contamination by large endogenous outward currents prevented an accurate measurement of V0.5,inact. However, 0.5 mM Cd2+ increased the amplitude of tail currents (Fig. 8B, D and F). This finding was consistent with a positive shift in the voltage dependence of inactivation in the absence of a significant effect on activation. Combined mutation of all three residues (D456A/D460A/D509A) further stabilized the closed state of the channel which could only be activated with very strong depolarizations. At these potentials, the large outward currents reflect activation of endogenous channels; however, the endogenous channels deactivate very rapidly compared to hERG (Fig. 8G) and do not significantly contaminate the tail currents used to measure the voltage dependence of activation. The estimated V0.5,act for the triple mutant channel was +125 mV and the 0.5 mM Cd2+-induced Vact was < 3 mV (Fig. 8H). Together, these findings suggest a crucial role for D456, D460 and D509 in the normal gating of hERG channels and that Cd2+ ions interact with all three residues in the WT channel.
, 百拇医药
A, currents conducted by D456A/D460A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from +30 to +110 mV, applied in 20-mV increments. Tail currents were measured at –70 mV. B, Cd2+ enhanced the tail current amplitude of D456A/D460A channels, but did not appreciably shift V0.5,act. C–F, the same protocols and analyses for D456A/D509A (C and D), D460A/D509A (E and F) and D456A/D460A/D509A (G and H) hERG channel currents. Examples of current traces for D456A/D509A and D460A/D509A (C and E) correspond to test potentials from 0 to +80 mV applied in 20-mV increments. D456A/D460A/D509A currents (G) were recorded in 20-mV increments to test potentials of +70 to +150 mV. Tail currents were measured at –20 mV.
, 百拇医药
Concentration-dependent effects of Cd2+ on hERG channel activation
Neutralization of D456 and D460 reduced Vact from +36 mV to +2.5 mV, suggesting that the high-affinity binding site for Cd2+ was effectively eliminated by this double mutation. However, these mutations were not expected to prevent higher concentrations of Cd2+ or other divalent cations from screening non-specific negative surface charges that exist near the extracellular regions of the channel. To differentiate between the gating effects caused by high-affinity binding to the putative S2–S3 pocket and screening of non-specific negative surface charge, the Vact induced by a range of [Cd2+]e(10 μM–30 mM) was determined for WT channels (Fig. 9A, ) and compared to D456A/D460A hERG channels (Fig. 9A, ). The Vact induced by 30 mM Cd2+ was +80 mV for WT channels and +23 mV for D456A/D460A channels, but even this high cation concentration did not cause a saturating effect. For the purpose of analysis, we assumed that the Vact induced by Cd2+ for D456A/D460A channels was caused by charge screening of non-specific negative surface charges and that a plot of the difference between the Vact–[Cd2+]e relationship for WT and D456A/D460A channels (Vact,subtract) versus [Cd2+]e was an estimate of the concentration–response relationship for Cd2+ binding to the S2–S3 site (Fig. 9B). The value of Vact,subtract saturated at 10 mM Cd2+. When fitted with a Hill equation, the maximum Vact,subtract was +58 mV, the EC50 for Cd2+ was 345 μM and the Hill coefficient was 0.59. The same subtraction and fitting procedure was repeated for the other mutant channels (Fig. 9C–F). Most mutations did not alter the concentration–response to Cd2+ (Table 3). As examples, the analyses for E518A and D591A mutant channels are shown in Fig. 9C. As expected, the D456A, D460A and D509A mutations reduced the efficacy (maximum response) and potency (EC50) of the effects of Cd2+ on channel activation (Fig. 9D). D509A had the greatest effect of the three mutations, reducing the maximum Vact,subtract to +16 mV and increasing the EC50 to 1.34 mM (Table 3). The double mutation of D460A/D509A was similar (data not shown), with a maximum Vact of +17 mV and an EC50 of 1.04 mM (Table 3).
, 百拇医药
A, Vact for WT hERG channels () and D456A/D460A channels (). Cd2+-induced Vact in this double mutant was considered to come from screening of non-specific surface charges. B, by subtracting Vact for D456A/D460A from Vact of WT channels we obtained an estimate of Cd2+-induced Vact (Vact,subtract) that results from specific binding to acidic residues in S2 and S3. The same procedure was applied to obtain the data presented in panels C–F. C, Vact,subtract for WT (), E518A () and D591A () channels. D, Vact,subtract for WT (), D456A (), D460A () and D509A () channels. E, Vact,subtractfor WT (), D411A () and D466A () channels. F, Vact,subtract for WT (), E519A () and C566A () channels. n = 4–13 for each concentration of Cd2+.
, http://www.100md.com
Unexpectedly, Vact,subtract for D411A and D466A channels was greater than WT channels when [Cd2+]e > 3 mM (Fig. 9E). These mutated residues are located near the cytosolic ends of the S1 and S2 domains in the hERG channel. The maximum Vact,subtract was +78 mV for D411A and +81 mV for D466A channels, an increase of about 20 mV relative to WT channels. In addition, both mutations caused a 2.75-fold decrease in Cd2+ potency (Table 3). Finally, two mutations (E519A and C566A) induced a small increase in Cd2+ potency without a significant change in maximum response (Fig. 9F and Table 3).
, 百拇医药
Ca2+ and Mg2+ interaction sites
The effects of 10 mM extracellular Ca2+ and 20 mM extracellular Mg2+ on Vact of WT and selected mutant hERG channels were determined. These concentrations are about 10-times greater than normal physiological concentrations. The effects of Ca2+ and Mg2+ were determined on the mutant channels that were most affected by Cd2+, including D411A, D456A, D460A, D466A, D509A, E518A and E519A. The Vact for WT channels was 26.1 mV for 10 mM Ca2+ (Table 4), a concentration about 100-times greater that the [Cd2+]e required for an equivalent shift in V0.5,act. The Ca2+-induced Vact was altered by mutation of D456, D460 and D466 in S2, D509 in S3, and E518A in the S3–S4 linker. The Vact for single and double mutants can be compared to estimate the relative importance of D456, D460 and D509 for binding of Ca2+. D456A/D460A channels were slightly less sensitive to Ca2+ than channels with either single mutation. D460A/D509A channels had the same Ca2+ sensitivity as mutation of D509 alone, whereas the combined effects of D456A and D509A were additive. Taken together these findings indicate that D456 and D509 are the most important residues with regard to Ca2+-induced alteration of hERG channel activation.
, 百拇医药
Mg2+ was less effective than Ca2+ in affecting hERG channel activation. A concentration of 20 mM Mg2+ was required for a Vact of +14.0 mV for WT channels. Unlike Ca2+, channel sensitivity to Mg2+ was not affected by D509A, and was enhanced by E518A (Table 4). D456A was the only mutation that significantly decreased the effects of Mg2+ on hERG channel gating, causing a reduction of Vact to 6.3 mV (Table 4). Double mutations of Asp residues in the S2 and S3 domains, especially D456/D460 and D456/D509, caused a synergistic decrease in the sensitivity to Mg2+. Although the functional consequences of single or double mutations differ for Cd2+, Ca2+ and Mg2+, our findings suggest that all three cations can interact with a coordination site formed by D456 and D460 residues in the S2 domain and D509 in the S3 domain.
, 百拇医药
Discussion
Effects of Cd2+ and other divalent cations on ERG channels
Extracellular divalent cations have multiple effects on voltage-gated ion channels. These cations can directly block the pore (Armstrong et al. 1982), screen negative surface charge and alter the apparent transmembrane voltage detected by intramembrane charged residues in the voltage sensors (McLaughlin, 1989), or bind to specific sites on the channel protein and alter the voltage dependence and kinetics of gating (Gilly & Armstrong, 1982; Armstrong & Lopez-Barneo, 1987; Begenisich, 1988). In this study we used site-directed mutagenesis to define the potential binding sites on the hERG channel that mediate the Cd2+-induced alterations in activation and inactivation gating. The obvious candidate residues involved with charge screening are acidic amino acids located in the extracellular linkers between the six transmembrane domains of each subunit. It is surprising that individual mutations of the four acidic amino acids in the S5–P linker to Ala did not significantly alter the ability of Cd2+ to shift V0.5,act. This finding is in contrast to the situation in Kv1–4 channels where the first five residues in the S5-P linker were proposed to be the main structural determinants of divalent cation-induced shifts of V0.5,act (Elinder et al. 1996, 1998). It was also surprising that combined neutralization of four negative charges in the S1–S2 linker did not cause a shift in the voltage dependence of hERG channel gating. Thus, the effects of relatively low concentrations (0.5 mM) of Cd2+ do not involve the 16 acidic residues located in the S1-S2 or the 16 acidic residues in the S5–P linkers present in a homotetrameric hERG channel.
, 百拇医药
Cd2+ slowed the activation, increased the rate of deactivation and induced positive shifts in the voltage dependence of activation and inactivation of hERG channel. In cardiac myocytes, ERG channels conduct the rapid delayed rectifier K+ current, IKr (Sanguinetti et al. 1995; Trudeau et al. 1995). The effects of Cd2+ on IKr was first studied by Follmer et al. (1992) who reported that 0.2 mM Cd2+ accelerated deactivation, shifted V0.5,act to more positive potentials and enhanced tail current magnitude in cat ventricular myocytes. The increase in current was attributed to a reduction in channel rectification. Similar effects of Cd2+ were reported for IKr recorded from guinea-pig myocytes (Daleau et al. 1997) and of Cd2+, Ni2+, Mn2+ and Co2+ in rabbit ventricular myocytes (Paquette et al. 1998). Co2+ was initially reported to block IKr (Sanguinetti & Jurkiewicz, 1991), but later experiments with rabbit IKr (Paquette et al. 1998) and hERG channels expressed in oocytes (Sanchez-Chapula & Sanguinetti, 2000) demonstrated that the decrease in current magnitude was caused by a positive shift in the voltage dependence of channel activation. Ca2+, Mg2+, Ba2+ and Sr2+ have similar, but much weaker effects than Co2+ or Cd2+ on IKr (Paquette et al. 1998). Although the effects of Ca2+ and Mg2+ on IKr or hERG channels was misinterpreted as a time- and voltage-dependent channel block that defines channel gating under normal physiological conditions (Ho et al. 1996, 1998; Song et al. 1999), some divalent cations can block hERG. For example, Zn2+ can block IKr in addition to modulating channel gating (Paquette et al. 1998).
, 百拇医药
Johnson et al. (1999) reported that a [Cd2+]e 0.2 mM caused a positive shift in the voltage dependence of rapid inactivation without affecting activation and that an inactivation-removed mutant hERG channel did not respond to 0.2 mM Cd2+. These findings were the basis of a model where low concentrations of Cd2+ had a direct effect on inactivation and that a lower affinity binding site and/or non-specific charge screening mediated the observed effects on activation gating. By contrast, we found that 0.5 mM Cd2+ slowed activation, accelerated deactivation and caused a +38 mV shift in V0.5,act of G628C/S631C hERG, effects that were similar to WT hERG channels. Nonetheless, as discussed below, some of our findings also support the idea that Cd2+ binds to two separate sites on the hERG channel.
, http://www.100md.com
Binding of Cd2+ to acidic residues in the S2 and S3 domains mediates the positive shift in hERG channel activation
External divalent cations can bind to specific sites on Kv channels to affect gating. This has been best studied in Shaker and dEAG channels by Papazian and colleagues (Tang et al. 2000; Silverman et al. 2000, 2004). In Shaker channels, acidic residues in S2 and S3 (E283, E293 and D316) are required for normal activation and channel folding and trafficking to the cell membrane (Tiwari-Woodruff et al. 1997). These effects are mediated by electrostatic interactions between these S2–S3 acidic residues and basic residues in S4 (Tiwari-Woodruff et al. 2000). In dEAG channels, neutralization of D278 in S2 and D327 in S3 slows channel activation and reduces or abolishes the slowing of activation caused by Mg2+ (Silverman et al. 2000, 2003, 2004). These findings suggest that Mg2+ (and Mn2+ and Ni2+) shield the negative side chains of D278 and D327 and prevent charge pairing with specific basic residues in the S4 domain. D278E abolishes Mg2+ but not Mn2+ sensitivity of EAG channels. The reason for this difference is unclear, but might be related to the ability of Mn2+ but not Mg2+ to be coordinated by a distorted octahedral geometry (Silverman et al. 2004). Coordination of Mg2+ by D278 and D327 in dEAG implies that these residues are positioned within a few angstroms of each other, constraining the structural arrangement of S2 and S3. Together with other findings, these constraints were used to construct a generalized structural model of the S1–S4 domains for Kv channels (Silverman et al. 2004) that differs markedly from the structure of the voltage sensor deduced by X-ray crystallography of the KvAP bacterial K+ channel (Jiang et al. 2003a,b).
, 百拇医药
Cd2+ binds to proteins in either a tetrahedral or octahedral arrangement and binding is most often mediated by interaction with the S of Cys, the N or N in the imidazole ring of His, or bidentate interactions with the O and N of the side chains of single Asp or Glu residues (Rulisek & Vondrasek, 1998). To identify the binding site for extracellular Cd2+ on the hERG channel, we mutated to Ala all potential coordination residues, including five Cys, three His, nine Asp and eight Glu residues located in transmembrane domains or extracellular loops linking these domains. Some of these mutations altered the normal gating of hERG channels and diminished the response to Cd2+.
, 百拇医药
Mutation of specific residues located in S2 (D456 and D460) or S3 (D509) domains greatly reduced the Cd2+-induced shift in V0.5,act. For example, the V0.5,act for D509A channels was shifted by +26 mV compared to the value in WT channels, but 0.5 mM Cd2+ only caused a further +8 mV shift. It could be argued that the ability of Cd2+ to shift V0.5,act was somehow impaired by mutations that altered the voltage dependence of activation. Evidence against this possibility is provided by two mutant channels. First, E518A channels also had a positive value for V0.5,act compared to WT channels, but responded to 0.5 mM Cd2+ with a +40 mV shift in V0.5,act, equivalent to that measured for WT channels. Second, the V0.5,act for D411A channels was 20 mV more negative than for WT channels, but 0.5 mM Cd2+ induced a normal +36-mV shift in this parameter. Thus, the response to Cd2+ was not correlated with the baseline value of V0.5,act. The site-directed mutagenesis approach is an indirect method to define residues important for ligand binding. An obvious limitation of the method is that it cannot distinguish between direct and allosteric effects. However, the specific mutations that most altered the response to Cd2+ were of residues that cluster together in relatively close proximity. This finding strongly suggests that D456, D460 and D509 form a binding site for Cd2+.
, 百拇医药
We were fortunate that most mutant channels functionally expressed in oocytes and retained relatively normal gating. This is in contrast to an earlier study where hERG channel residues were mutated to Cys and it was found that D456C was non-functional. Furthermore, because substitution of other S2 and S3 Asp residues with Cys did not alter the Ca2+-induced shift in V0.5,act of hERG, it was concluded that these Asp residues did not form a metal ion-binding site (Liu et al. 2003). Because Cys might adequately substitute for one of the Asp residues in coordination of Ca2+, Cys-substitution may not be a valid test of the ability of a specific residue to cooperate in divalent cation binding. In our study, D456 was mutated to Ala. D456A channels expressed well in oocytes and this mutation interfered with the Cd2+-induced shift in V0.5,act, as did mutations of other S2 and S3 acidic residues.
, http://www.100md.com
Combined mutation of multiple acidic residues provided further evidence for a Cd2+ coordination site in a pocket formed by the S2 and S3 domains. The shift in V0.5,act for D456A/D509A channels was equivalent to the sum of the shifts induced by individual mutations, whereas D456A/D460A and D460A/D509A channels were shifted more than expected from the sum of the shift of the single mutations. This finding suggests that the functional effects of mutations of D456 or D509 are independent and that efficient coordination of V0.5,act Cd2+ in the putative pocket formed between S2 and S3 is most dependent on these two residues. The Cd2+-induced shift in activation gating that remained after mutation of both D456 and D460 probably results from screening of sialic acid, the typical terminal residue on cell surface oligosaccharides and other negatively charged molecules bound to the extracellular cell surface.
, 百拇医药
Charge pairing between acidic residues in S2–S3 and basic residues in S4 can serve to promote or stabilize voltage sensor movement and/or facilitate correct protein folding and trafficking of the channel complex to the cell surface. Our finding that the triple mutant (D456A/D460A/D509A) functionally expressed indicates that charge pairing between basic residues of S4 and these acidic residues is not an absolute requirement for protein folding, trafficking or gating.
, http://www.100md.com
Cd2+ is commonly coordinated in a tetrahedral arrangement in metalloproteins (Rulisek & Vondrasek, 1998). It is possible that extracellular Cd2+ may also be coordinated in a tetrahedral arrangement in a pocket between S2 and S3 domains of each hERG subunit and that three of the coordination residues are D456, D460 and D509. The identity of the putative fourth coordination point is unknown, but could be a water molecule or a peptidyl oxygen of another unidentified residue. Double mutations of D456, D460 or D509 suggest that these three residues also contribute to a lower-affinity binding site for Ca2+ and Mg2+ and that D460 is the least important of the three residues for coordinating divalent cations. It seems likely that all divalent cations that affect activation gating of hERG channel, including Ca2+, Mg2+, Ba2+, Sr2+, Mn2+, Co2+, Cd2+, Zn2+, and Ni2+ (Follmer et al. 1992; Paquette et al. 1998; Ho et al. 1998, 1999; Johnson et al. 1999, 2001; Sanchez-Chapula & Sanguinetti, 2000), bind to the same S2/S3 site.
, 百拇医药
Coupling between activation and inactivation
For some residues, the mutation-induced shift in V0.5,act and V0.5,inact were the same. For example, D456A shifted V0.5,act by +33 mV and V0.5,inact by +29 mV. For some residues, Ala substitution had a differential effect on the voltage dependence of activation and inactivation. For example, D411A shifted V0.5,act by –19 mV, but shifted V0.5,inact by +14 mV. D460A and D509A shifted V0.5,act by +22 and +27 mV, but had no significant effect on V0.5,inact. We have previously noted a disconnection between the effects of a point mutation on activation and inactivation gating. S631A shifts V0.5,inact by +100 mV but has no effect on V0.5,act (Zou et al. 1998) and R531A shifts V0.5,act by +57 mV without affecting V0.5,inact (Piper et al. 2005). As shown in the present study, extracellular cations can also differentially affect gating associated with activation and inactivation. For example, 0.5 mM Cd2+ shifted V0.5,act of WT hERG channel by +36 mV, twice as much as the shift in V0.5,inact (+18 mV). Thus, shifts in the voltage dependence of inactivation induced by mutation or divalent cations can differ significantly from the voltage dependence of activation. This difference could be caused by alteration of the coupling between activation and inactivation and does not necessarily imply that the two gating processes are controlled by different voltage sensors (Piper et al. 2003).
, 百拇医药
The binding site for the Cd2+-mediated shift in hERG channel inactivation remains elusive
Cd2+ destabilizes the inactivated state of the hERG channel. A poorly defined rearrangement of the S5-P linker/pore helix is presumed to be responsible for rapid C-type inactivation (Smith et al. 1996; Herzberg et al. 1998; Ficker et al. 2001; Liu et al. 2002). The S5-P linker of hERG channel is 42 amino acids in length and considerably longer than most other Kv channels that have a S5-P linker length of < 15 residues. [2-(trimetylammonium) ethyl] methane sulfonate (MTSET) modification of Cys residues introduced into the middle of the hERG linker (residues 583–597) disrupted channel inactivation and this region was proposed to form an -helix that bridged between the voltage sensor (S4) and the outer mouth of the channel pore (Liu et al. 2002). In contrast, modification of residues closer to S5 (residues 574–582) by MTSET had only small effects on gating and were proposed to face away from the channel and towards the extracellular aqueous solution. The structure of the isolated S5-P linker, determined by NMR spectroscopy, was disordered in an aqueous solution (Torres et al. 2003). However, in the presence of micelles, the linker formed two helical regions (Trp585–Ile593 and Gly604–Tyr611). The first helix was preceded by a highly flexible stretch of amino acids (from Ala571 to Gly585) (Torres et al. 2003) that has previously been reported to be little affected by thiol modification of introduced Cys residues (Liu et al. 2002). Mutation of E575, H578 or D580 to Ala partially inhibited the effects of Cd2+ on inactivation. Based on NMR findings, these three residues are located in a disordered region of the S5-P linker that is readily accessible by extracellular Cd2+.
, http://www.100md.com
Two findings suggest that the effect of Cd2+ on inactivation is mediated by binding to a site distinct from the S2–S3 pocket. First, mutation analysis of D456, D460 and D509 indicates that Cd2+ can shift V0.5,inact independently of its effects on activation gating. Second, mutation of E575, H578 or D580 reduced the Cd2+-induced shift in V0.5,inact, but did not alter the Cd2+-induced shift in V0.5,act. Ala substitution of these three residues did not alter V0.5,inact in the absence of Cd2+, suggesting that these specific residues are not crucial for normal coupling between activation and inactivation. However, combined mutations of all three residues did not prevent the Cd2+-induced shift in V0.5,inact as would be predicted if these residues formed a binding pocket. Cd2+ may bind to another unidentified site to affect inactivation gating.
, 百拇医药
Summary
Mutation of three acidic residues in S2 and S3 of the hERG channel (D456, D460 and D509) greatly attenuated the extracellular Cd2+-induced positive shifts in the voltage dependence of activation but not inactivation. We conclude that coordination of Cd2+ by these acidic residues prevent specific charge pair interactions with basic residues in S4 that normally facilitates hERG channel gating associated with channel activation.
, http://www.100md.com
References
Anumonwo JM, Horta J, Delmar M, Taffet SM & Jalife J (1999). Proton and zinc effects on HERG currents. Biophys J 77, 282–298.
Armstrong CM & J (1987). External calcium ions are required for potassium channel gating in squid neurons. Science 236, 712–714.
Armstrong CM, Swenson RP Jr & Taylor SR (1982). Block of squid axon K channels by internally and externally applied barium ions. J Gen Physiol 80, 663–682.
, 百拇医药
Begenisich T (1988). The role of divalent cations in potassium channels. Trends Neurosci 11, 270–273.
Cherny VV & DeCoursey TE (1999). pH-dependent inhibition of voltage-gated H+ currents in rat alveolar epithelial cells by Zn2+ and other divalent cations. J Gen Physiol 114, 819–838.
Daleau P, Khalifa M & Turgeon J (1997). Effects of cadmium and nisoldipine on the delayed rectifier potassim current in guinea pig ventricular myocytes. J Pharmacol Exp Ther 281, 826–833.
, 百拇医药
Elinder F, Liu Y & Arhem P (1998). Divalent cation effects on the Shaker K channel suggest a pentapeptide sequence as determinant of functional surface charge density. J Membr Biol 165, 183–189.
Elinder F, Madeja M & Arhem P (1996). Surface charges of K channels. Effects of strontium on five cloned channels expressed in Xenopus oocytes. J Gen Physiol 108, 325–332.
Ficker E, Jarolimek W & Brown AM (2001). Molecular determinants of inactivation and dofetilide block in ether a-go-go (EAG) channels and EAG-related K+ channels. Mol Pharmacol 60, 1343–1348.
, 百拇医药
Follmer CH, Lodge NJ, Cullinan CA & Colatsky TJ (1992). Modulation of the delayed rectifier, IK, by cadmium in cat ventricular myocytes. Am J Physiol 262, C75–C83.
Gilly WF & Armstrong CM (1982). Divalent cations and the activation kinetics of potassium channels in squid giant axons. J Gen Physiol 79, 965–996.
Herzberg IM, Trudeau MC & Robertson GA (1998). Transfer of rapid inactivation and sensitivity to the class III antiarrhythmic drug E-4031 from HERG to M-eag channels. J Physiol 511, 3–14.
, http://www.100md.com
Ho W-K, Earm YE, Lee SH, Brown HF & Noble D (1996). Voltage- and time-dependent block of delayed rectifier K+ current in rabbit sino-atrial node cells by external Ca2+ and Mg2+. J Physiol 494, 727–742.
Ho W-K, Kim I, Lee CO & Earm YE (1998). Voltage-dependent blockade of HERG channels expressed in Xenopus oocytes by external Ca2+ and Mg2+. J Physiol 507, 631–638.
Ho W-K, Kim I, Lee CO, Youm JB, Lee SH & Earm YE (1999). Blockade of HERG channels expressed in Xenopus laevis oocytes by external divalent cations. Biophys J 76, 1959–1971.
, 百拇医药
Jiang Y, Lee A, Chen J, Ruta V, Cadene M, Chait BT & MacKinnon R (2003a). X-ray structure of a voltage-dependent K+ channel. Nature 423, 33–41.
Jiang Y, Ruta V, Chen J, Lee A & MacKinnon R (2003b). The principle of gating charge movement in a voltage-dependent K+ channel. Nature 423, 42–48.
Johnson JP Jr, Balser JR & Bennett PB (1999). Enhancement of HERG K+ currents by Cd2+ destabilization of the inactivated state. Biophys J 77, 2534–2541.
, http://www.100md.com
Johnson JP Jr, Balser JR & Bennett PB (2001). A novel extracellular calcium sensing mechanism in voltage-gated potassium ion channels. J Neurosci 21, 4143–4153.
Keating MT & Sanguinetti MC (2001). Molecular and cellular mechanisms of cardiac arrhythmias. Cell 104, 569–580.
Liu J, Zhang M, Jiang M & Tseng GN (2002). Structural and functional role of the extracellular S5-P linker in the HERG potassium channel. J Gen Physiol 120, 723–737.
, 百拇医药
Liu J, Zhang M, Jiang M & Tseng GN (2003). Negative charges in the transmembrane domains of the HERG K channel are involved in the activation- and deactivation-gating processes. J Gen Physiol 121, 599–614.
McLaughlin S (1989). The electrostatic properties of membranes. Annu Rev Biophys Biophys Chem 18, 113–136.
Mitcheson JS, Chen J, Lin M, Culberson C & Sanguinetti MC (2000). A structural basis for drug-induced long QT syndrome. Proc Natl Acad Sci U S A 97, 12329–12333.
, 百拇医药
Papazian DM, Shao XM, Sech S-A, Mock AF, Huang Y & Wainstock DH (1995). Electrostatic interactions of S4 voltage sensor in Shaker K+ channel. Neuron 14, 1293–1301.
Paquette T, Clay JR, Ogbaghebriel A & Shrier A (1998). Effects of divalent cations on the E-4031-sensitive repolarization current, IKr, in rabbit ventricular myocytes. Biophys J 74, 1278–1285.
Piper DR, Hinz WA, Tallurri CK, Sanguinetti MC & Tristani-Firouzi M (2005). Regional specificity of human ether-à-go-go-related gene channel activation and inactivation gating. J Biol Chem 280, 7206–7217.
, 百拇医药
Piper DR, Varghese A, Sanguinetti MC & Tristani-Firouzi M (2003). Gating currents associated with intramembrane charge displacement in HERG potassium channels. Proc Natl Acad Sci U S A 100, 10534–10539.
Rulisek L & Vondrasek J (1998). Coordination geometries of selected transition metal ions (Co2+, Ni2+, Cu2+, Zn2+, Cd2+, and Hg2+) in metalloproteins. J Inorg Biochem 71, 115–127.
Sanchez-Chapula JA & Sanguinetti MC (2000). Altered gating of HERG potassium channels by cobalt and lanthanum. Pflugers Arch 440, 264–274.
, 百拇医药
Sanguinetti MC, Jiang C, Curran ME & Keating MT (1995). A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell 81, 299–307.
Sanguinetti MC & Jurkiewicz NK (1991). IK is comprised of two components in guinea pig atrial cells. Am J Physiol 260, H393–H399.
Sanguinetti MC & Xu QP (1999). Mutations of the S4–S5 linker alter activation properties of HERG potassium channels expressed in Xenopus oocytes. J Physiol 514, 667–675.
, 百拇医药
Seoh S-A, Sigg D, Papazian DM & Bezanilla F (1996). Voltage-sensing residues in the S2 and S4 segments of the Shaker K+ channel. Neuron 16, 1159–1167.
Silverman WR, Bannister JP & Papazian DM (2004). Binding site in Eag voltage sensor accommodates a variety of ions and is accessible in closed channel. Biophys J 87, 3110–3121.
Silverman WR, Roux B & Papazian DM (2003). Structural basis of two-stage voltage-dependent activation in K+ channels. Proc Natl Acad Sci U S A 100, 2935–2940.
, 百拇医药
Silverman WR, Tang CY, Mock AF, Huh KB & Papazian DM (2000). Mg2+ modulates voltage-dependent activation in ether-a-go-go potassium channels by binding between transmembrane segments S2 and S3. J Gen Physiol 116, 663–678.
Smith PL, Baukrowitz T & Yellen G (1996). The inward rectification mechanism of the HERG cardiac potassium channel. Nature 379, 833–836.
Song DK, Earm YE & Ho WK (1999). Blockade of the delayed rectifier K+ currents, IKr, in rabbit sinoatrial node cells by external divalent cations. Pflugers Arch 438, 147–153.
, 百拇医药
Spector PS, Curran ME, Zou A, Keating MT & Sanguinetti MC (1996). Fast inactivation causes rectification of the IKr channel. J Gen Physiol 107, 611–619.
Stuhmer W (1992). Electrophysiological recording from Xenopus oocytes. Methods Enzymol 207, 319–339.
Tang CY, Bezanilla F & Papazian DM (2000). Extracellular Mg2+ modulates slow gating transitions and the opening of Drosophila ether-a-Go-Go potassium channels. J Gen Physiol 115, 319–338.
, http://www.100md.com
Tiwari-Woodruff SK, Lin MA, Schulteis CT & Papazian DM (2000). Voltage–dependent structural interactions in the shaker K+ channel. J Gen Physiol 115, 123–138.
Tiwari-Woodruff SK, Schulteis CT, Mock AF & Papazian DM (1997). Electrostatic interactions between transmembrane segments mediate folding of Shaker K+ channel subunits. Biophys J 72, 1489–1500.
Torres AM, Bansal PS, Sunde M, Clarke CE, Bursill JA, Smith DJ, Bauskin A, Breit SN, Campbell TJ, Alewood PF, Kuchel PW & Vandenberg JI (2003). Structure of the HERG K+ channel S5P extracellular linker: role of an amphipathic alpha-helix in C-type inactivation. J Biol Chem 278, 42136–42148.
, http://www.100md.com
Trudeau M, Warmke JW, Ganetzky B & Robertson GA (1995). HERG, a human inward rectifier in the voltage-gated potassium channel family. Science 269, 92–95.
Warmke JW & Ganetzky B (1994). A family of potassium channel genes related to eag in Drosophila and mammals. Proc Natl Acad Sci U S A 91, 3438–3442.
Zou A, Xu QP & Sanguinetti MC (1998). A mutation in the pore region of HERG K+ channels reduces rectification by shifting the voltage dependence of inactivation. J Physiol 509, 129–138., 百拇医药(David Fernandez, Azad Gha)
Abstract
Cd2+ slows the rate of activation, accelerates the rate of deactivation and shifts the half-points of voltage-dependent activation (V0.5,act) and inactivation (V0.5,inact) of human ether-à-go-go-related gene (hERG) K+ channels. To identify specific Cd2+-binding sites on the hERG channel, we mutated potential Cd2+-coordination residues located in the transmembrane domains or extracellular loops linking these domains, including five Cys, three His, nine Asp and eight Glu residues. Each residue was individually substituted with Ala and the resulting mutant channels heterologously expressed in Xenopus oocytes and their biophysical properties determined with standard two-microelectrode voltage-clamp technique. Cd2+ at 0.5 mM caused a +36 mV shift of V0.5,act and a +18 mV shift of V0.5,inact in wild-type channels. Most mutant channels had a similar sensitivity to 0.5 mM Cd2+. Mutation of single Asp residues located in the S2 (D456, D460) or S3 (D509) domains reduced the Cd2+-induced shift in V0.5,act, but not V0.5,inact. Combined mutations of two or three of these key Asp residues nearly eliminated the shift induced by 0.5 mM Cd2+. Mutation of D456, D460 and D509 also reduced the comparatively low-affinity effects of Ca2+ and Mg2+ on V0.5,act. Extracellular Cd2+ modulates hERG channel activation by binding to a coordination site formed, at least in part, by three Asp residues.
, http://www.100md.com
Introduction
Human ether-à-go-go-related gene (hERG) channel is a member of the ether-à-go-go (EAG) family of voltage-gated K+ channels (Warmke & Ganetzky, 1994) and is expressed in many cell types, including cardiac myocytes, neurones and tumour cells. In cardiomyocytes, hERG channel subunits co-assemble to form homotetrameric channels that conduct the rapid delayed rectifier K+ current IKr (Sanguinetti et al. 1995; Trudeau et al. 1995). IKr is an important component of the outward currents that mediate repolarization of the cardiac action potential. Inherited loss-of-function mutations in HERG or block of hERG channels by specific drugs prolongs the duration of action potentials and causes long QT syndrome, a disorder of ventricular repolarization that increases the risk of life-threatening arrhythmias (Keating & Sanguinetti, 2001).
, 百拇医药
Divalent cations can modify ion channel gating by screening non-specific negative surface charges located on membrane lipids and proteins (McLaughlin, 1989) or by binding to specific residues of channel proteins (Gilly & Armstrong, 1982). Recent studies have attempted to localize the binding sites using site-directed mutagenesis. Several acidic residues in the S5–P linker of Kv1–Kv4 channels were proposed to account for the majority of negative surface charges that were screened by divalent cations (Elinder et al. 1996, 1998). In Drosophila EAG (dEAG) channels, Mg2+, Mn2+ and Ni2+ slow gating by binding to an extracellular site formed by two acidic residues: D278 in S2 and D327 in S3 (Silverman et al. 2000). Charge pair interactions between these Asp residues and specific basic residues in S4 have been proposed to facilitate activation of dEAG channels and protein folding and activation of Shaker channels (Papazian et al. 1995; Seoh et al. 1996; Tiwari-Woodruff et al. 1997, 2000; Silverman et al. 2000, 2003, 2004). Specific charge pairings may be disrupted and replaced by others as the S4 helix slides past the relatively immobile S2 and S3 domains. In this scheme, divalent cations bind to the Asp residues in S2 or S3 and prevent interaction with S4, causing a slowing of EAG channel activation.
, http://www.100md.com
Only a few of the many residues in hERG channel that could potentially interact with divalent cations have been investigated and there is no consensus regarding the location of a binding site. Asp residues in S2 and S3 of dEAG are conserved in hERG. D278, D282 and D327 of dEAG are homologous to D456, D460 and D509 of hERG channel (Fig. 1A)
A, sequence alignment of S2 and S3 domains of hERG and EAG channels. B, topology of a single hERG channel subunit with six transmembrane domains (S1–S6) that span the membrane. The approximate location of the Asp, Glu, His and Cys residues mutated to Ala in this study are shown. The residues that interact with Cd2+ are highlighted in large italic text.
, 百拇医药
The conservation of charged residues in S2 and S3 suggests that divalent cations could bind to a common site on dEAG and hERG channels and affect gating in a similar manner. However, several recent studies argue against these simple predictions. First, Mg2+ greatly slows the rate of dEAG channel opening, but unlike in hERG, does not cause a shift in the voltage dependence of activation (Tang et al. 2000). Second, mutation of the acidic residues of S2 and S3 to Cys did not alter Ca2+-induced changes in hERG channel gating (Liu et al. 2003). Third, the shift in hERG gating by Ca2+ is influenced by an acidic residue (E519) located in the extracellular S3–S4 linker and neutralization of an adjacent residue (E518) shifts the voltage dependence of activation to more positive potentials and increases the affinity of the channel for Ca2+ (Johnson et al. 2001). Fourth, more than one binding site may account for the multiple effects of divalent cations on hERG or dEAG channel gating. For example, mutation of D327 to Ala in dEAG eliminates Mg2+ but not Ni2+ sensitivity, suggesting that the coordination sites for Ca2+ and Ni2+ overlap, but are not identical (Silverman et al. 2004). The structural basis of divalent cation-mediated changes in hERG channel gating is unclear and a more systematic study is warranted.
, http://www.100md.com
Alkaline earth cations (e.g. Ca2+ and Mg2+) slow the rate of activation, accelerate the rate of deactivation and induce positive shifts in the voltage dependence of hERG channel activation, and to a lesser extent, inactivation (Ho et al. 1996, 1998; Johnson et al. 1999, 2001; Silverman et al. 2000, 2003). Transition metal ions (e.g. Cd2+, Co2+, Ni2+ and Zn2+) have similar effects on hERG channel gating, but are more potent (Anumonwo et al. 1999; Johnson et al. 1999; Sanchez-Chapula & Sanguinetti, 2000). Previous reports have suggested that divalent cations may differentially affect hERG channel activation and inactivation. For example, low concentrations of Cd2+ (< 200 μM) preferentially shifted the voltage dependence of hERG channel inactivation, causing reduced rectification of the current–voltage (I–V) relationship and an increased peak outward current, whereas higher Cd2+ concentrations were required to affect activation (Johnson et al. 1999).
, http://www.100md.com
The coordination geometry of transition metal ions can be complex. Based on a comprehensive survey of binding residues in metalloproteins (Rulisek & Vondrasek, 1998), the typical coordination geometry of Cd2+ is either tetrahedral or octahedral and can involve up to five different amino acid residues with the following relative order of importance: Cys > Asp > His > Glu > Asn. Most often, one or two sites in the tetrahedral, and more than two sites in the octahedral, coordination complex are water molecules. To identify Cd2+-binding sites on hERG channel, it is important to assay all potential binding residues because the 3-dimensional structure of the channel is incompletely understood. Towards this goal, we have mutated all 25 amino acids that could conceivably form a coordination site for extracellular Cd2+, including nine Asp, eight Glu, three His and five Cys residues (Fig. 1B). Residues were mutated to Ala and the resulting mutant channels expressed in Xenopus oocytes. The voltage dependence and kinetics of activation and inactivation of mutant channels were compared to wild-type (WT) hERG channels using two-microelectrode voltage-clamp technique. Mutation of three Asp residues in S2 and S3 caused a positive shift in V0.5,act and selectively prevented further shifts normally induced by extracellular Cd2+. Mutation of three amino acids located in the S5–pore linker only affected the Cd2+-induced shift in V0.5,inact. Together these findings suggest that Cd2+ binds to different sites on hERG channel to differentially affect channel activation and inactivation.
, http://www.100md.com
Methods
Molecular biology
hERG channel was subcloned into the pSP64 plasmid expression vector (Promega, Madison, WI, USA) and site-directed mutagenesis was performed as previously described (Mitcheson et al. 2000). Mutations were confirmed by restriction mapping and DNA sequencing. Complementary RNA (cRNA) for injection into oocytes was prepared with SP6 Cap-Scribe (Roche Applied Science, Indianapolis, IN, USA) following linearization of the expression construct with EcoRI.
, http://www.100md.com
Oocytes and solutions
Oocytes were isolated by dissection from anaesthetized adult Xenopus laevis. Frogs were anaesthetized by immersion in 0.2% tricaine (Sigma) for 10–15 min. A small abdominal incision was made and ovarian lobes containing oocytes were removed. The incision was sutured closed and the frog returned to its aquarium for a recovery period of at least 1 month before the procedure was repeated. After a maximum of three surgical procedures, tricaine-anaesthetized frogs were killed by pithing. Clusters of oocytes were treated with 2 mg l–1 type 2 collagenase (Worthington) to remove follicle cells. Maintenance and cRNA injections into oocytes were performed as previously described (Stuhmer, 1992; Sanguinetti & Xu, 1999). During voltage-clamp experiments, the oocytes were bathed in a solution containing, (mM): NaCl 96, KCl 2, CaCl2 1, MgCl2 2 and Hepes 5; pH adjusted to 7.6 with NaOH. Stock solutions of CdCl2 (Sigma) were prepared in this solution at concentrations of 0.1 or 1 M and kept at 4°C. CdCl2 remained soluble in the saline solution at 30 mM, the highest concentration used in this study. No compensation was made for changes in extracellular osmolarity that resulted from addition of CdCl2 to the bathing solution or for Cd2+ binding to Hepes (Cherny & DeCoursey, 1999). The solution flow rate was 1.5 ml min–1 and the oocyte chamber volume was 0.2 ml.
, 百拇医药
Voltage-clamp and data analysis
Ionic currents were recorded at room temperature (22–24°C) using a GeneClamp 500 amplifier (Molecular Devices, Sunnyvale, CA, USA) and standard two-microelectrode voltage-clamp technique (Stuhmer, 1992). Isochronal I–V relationships were determined with 2-s test pulses applied at a frequency of 0.06 Hz and in 10-mV increments to voltages ranging from –70 mV up to +120 mV. The holding potential was –80 mV. The amplitude of current measured at the end of the 2-s pulse was normalized to the peak value measured under control conditions (before exposure of an oocyte to CdCl2).
, http://www.100md.com
Fully activated I–V relationships were determined from analysis of tail currents. A 2-s prepulse to a voltage where activation was complete for each mutant channel and condition (indicated in figure legends) was applied before repolarization to a variable test potential. Tail currents during the test potential were fitted to a bi-exponential function using the Chebyshev method for curve fitting provided by Clampfit 8 software (Molecular Devices, Sunnyvale, CA). The peak tail current (Itail) amplitude was determined by extrapolation of fitted current traces back to the moment of membrane repolarization. Itail for each pulse was normalized to the largest value (Itail,max) measured under control conditions and plotted as a function of test voltage (Vt). The kinetics of channel deactivation were quantified by time constants and relative amplitudes of the fast and slow components of tail current decay.
, http://www.100md.com
The conductance–voltage (g–V) relationship for hERG channel current was determined from peak Itail elicited after 2-s pulses to a variable test potential. Itail for each test potential was normalized to Itail,max. The resulting relative conductance (g/gmax) values were plotted against Vt, and the relationship fitted to a Boltzmann function (eqn (1)) using Origin 7 software (OriginLab, Northampton, MA, USA).
where V0.5,act is the voltage required to elicit 0.5 x g/gmax and k is the slope factor of the relationship.
, 百拇医药
The voltage dependence for the recovery from hERG channel inactivation was determined with a three-pulse protocol (Smith et al. 1996; Spector et al. 1996). First, hERG channels were activated or inactivated by a 1-s prepulse to a potential (Vpre) that was varied from +40 to +90 mV depending on the specific mutation. In a second pulse, channels were allowed to recover from inactivation during a 10-ms pulse applied to a test voltage (Vt) that was increased in 10-mV steps from –140 mV to the voltage used for the first pulse. During the third and final pulse, channels were inactivated by stepping the membrane potential back to Vpre and the resulting ionic current following the decay of the capacitance current was fitted with a mono-exponential function using the Levenberg-Marquardt method in pCLAMP software (Molecular Devices). The amplitude of inactivating current (Iinact) was determined by extrapolation of the fitted current trace back to the initial time of the third pulse. Iinact was normalized to the largest value (Iinact,max) and plotted as a function of Vt. Iinact – Vt curves were fitted with a Boltzmann function similar to eqn (1) to estimate V0.5,inact and k for the relationship. In this study, we use the term ‘inactivation’ to refer to the voltage dependence of recovery from inactivation.
, 百拇医药
The initial screening of mutant channels utilized an extracellular Cd2+ concentration ([Cd2+]e) of 0.5 mM. Some of the mutant channels were further characterized by determining the effect of 10 μM to 30 mM [Cd2+]e. Concentration–effect data were fitted to the Hill equation (eqn (2)) to determine the EC50, the effective [Cd2+]e required for the shift in V0.5,act (Vact) to reach half its maximum value (Vmax), and the Hill coefficient (h).
Data in tables are presented as mean ± S.E.M. (n = number of cells) and statistical comparisons between experimental groups were performed using the Student's t test. Differences were considered significant at P < 0.05.
, 百拇医药
Results
Effects of Cd2+ on wild-type hERG channels
We first determined the effects of a single concentration of Cd2+ (0.5 mM) on hERG channels expressed in Xenopus oocytes. Extracellular application of 0.5 mM Cd2+ caused an immediate change in the kinetics and amplitude of WT hERG channel currents elicited by a 2-s pulse to +60 mV from a holding potential of –80 mV (Fig. 2A). The most notable effects were a slowing of activation (Fig. 2B) and accelerated deactivation at –70 mV (Fig. 2C). These effects were rapidly and completely reversible upon washout of Cd2+ from the bathing solution (not shown), indicating an extracellular site of action. Currents elicited in response to 2-s test potentials ranging from –70 to +70 mV are shown in Fig. 2D. Currents measured at the end of each pulse were normalized relative to the peak values at –10 mV under control conditions and the averages for 11 oocytes plotted as a function of test potential in Fig. 2E. Cd2+ increased current magnitudes for test potentials positive to +20 mV, but decreased currents at test potentials negative to +20 mV. Cd2+ also shifted the peak of the I–V relationship by 35 mV, equivalent to the positive shift in the voltage dependence of activation (V0.5,act, +36.4 ± 0.8, n = 11) determined by plotting the normalized tail current amplitude as a function of test potential (Fig. 2F).
, 百拇医药
A, currents recorded during 2-s pulses to +60 mV before (thin trace) and in the presence (thick trace) of 0.5 mM Cd2+. Tail currents were measured at –70 mV. B and C, the currents inside the dashed boxes in A are shown in an expanded view. D, currents recorded at test potentials of –50 to +50 mV, applied in 20-mV increments before (left) and after (right) exposure to Cd2+. E, normalized isochronal I–V relationships for currents measured at end of 2-s pulses before () and after application of Cd2+ () (n = 11). F, activation curves determined from peak tail currents following 2-s pulses to the indicated potentials. G, currents recorded before (left) and after (right) using a two-pulse voltage-clamp protocol shown in inset. To ensure currents were fully activated, the first pulse was to +30 mV in the absence of Cd2+ (left) and +60 mV after adding 0.5 mM Cd2+ (right). Current traces correspond to test potentials of –120 to +30 mV applied in 30-mV increments. H, fully activated I–V relationships for peak (extrapolated) tail currents using protocol shown in G. I, relative amplitude of fast component of deactivating current. J and K, time constants for current deactivation. Symbols in F and H–K are the same as in E.
, 百拇医药
The kinetics of hERG channel current deactivation were determined over a range of membrane potentials. The oocyte was first depolarized to +60 mV for 2 s before repolarization to a potential that was varied from +30 to –140 mV. Examples of currents recorded with this pulse regimen before, and 3 min after, addition of 0.5 mM Cd2+ to the extracellular solution are shown in Fig. 2G. The peak currents during the second pulse were plotted as a function of voltage to obtain the ‘fully activated’ I–V relationship (Fig. 2H). These I–V relationships exhibit marked inward rectification because channels only partially recover from rapid C-type inactivation at potentials positive to –100 mV. Tail currents were fitted with a bi-exponential function to determine the relative amplitudes (Fig. 2I), and the time constants of the fast and slow components of current decay (Fig. 2J and K). Cd2+ accelerated the rate of deactivation at all potentials examined and caused more than a +30-mV shift in the voltage dependence of these measures of deactivation, similar to the Cd2+-induced shift in V0.5,act. The increase in outward currents by Cd2+ at potentials positive to –60 mV in the fully activated I–V relationship (Fig. 2H) suggested an effect on inactivation gating. Therefore, the effects of 0.5 mM Cd2+ on the voltage dependence of recovery from C-type inactivation were determined with a three-pulse protocol shown in Fig. 3A (Smith et al. 1996; Spector et al. 1996). Cd2+ shifted the half-point for this relationship (V0.5,inact) by +18 mV (Fig. 3B), a shift that was half the Cd2+-induced effect on channel activation. The effect of Cd2+ on activation gating caused a shift in the peak of the I–V relationship from –10 to +20 mV. The lesser effect of Cd2+ on inactivation gating reduced the peak of the I–V relationship with a cross-over at potentials > +20 mV (Fig. 2E).
, http://www.100md.com
A, representative current recordings of WT hERG channel before (left) and after (right) exposure to 0.5 mM Cd2+. Currents shown were elicited by pulsing to test potentials from –130 to –30 mV applied in 20-mV increments after a 1-s prepulse to +40 mV before and +60 mV in the presence of Cd2+ (not shown). Inset shows the voltage changes during the second of the three pulses. B, plot of fractional availability of channels to open (relative amplitude of extrapolated currents elicited during third step of the three-pulse protocol) in the absence () and presence of 0.5 mM Cd2+ (). The V0.5,inact was –89.4 mV and k was 22.1 mV for control and –71.4 mV and 22.4 mV, respectively, after exposure of oocytes to 0.5 mM Cd2+. C, Cd2+ binds to and modifies gating of inactivated hERG channels. Currents were recorded during 40-s pulses to +50 mV before (trace i), during (trace ii) and after (trace iii) exposure of the oocyte to 0.5 mM Cd2+. The solution switch was made at the time indicated by the arrow during trace ii. The time between pulses was 15 s.
, http://www.100md.com
The rate of hERG channel activation was slower during the first depolarizing pulse applied immediately after switching to a bathing solution containing Cd2+. This finding suggests that Cd2+ can bind to the closed state of the channel to affect gating associated with a subsequent voltage-dependent activation. To determine whether modification of gating can occur when the channel is in other states, we switched the control bathing solution to one containing 0.5 mM Cd2+ during a prolonged (40 s) voltage-clamp pulse to +50 mV, a voltage where all channels were predicted to be in either the open or inactivated state. In the example shown in Fig. 3C, the solution switch was made 10 s after the onset of the 40-s pulse. The current amplitude increased immediately (Fig. 3C, trace ii), then slowly enhanced over the next 30 s to a new steady-state level. The increase in current magnitude at +50 mV was consistent with some channels recovering from an inactivated state to an open state and suggests that Cd2+ destabilized the inactivated channel state.
, 百拇医药
Effects of Cd2+ on inactivation-deficient hERG channels
The double mutation G628C/S631C eliminates fast inactivation of the hERG channel (Smith et al. 1996). At low concentrations, Cd2+ was reported to have only minor and insignificant effects on current amplitude or kinetics of deactivation of this mutant channel, suggesting that channel inactivation might be essential for a Cd2+-induced increase in hERG channel current (Johnson et al. 1999). We determined the effects of 0.5 mM Cd2+ on G628C/S631C hERG channels. Cd2+ increased current magnitude at potentials positive to +20 mV (Fig. 4A and B). The mechanism of this effect is unknown, but it is possible that Cd2+ might bind to these introduced Cys residues and affect pore properties (e.g. increased single channel conductance or mean open time). Cd2+ (0.5 mM) shifted V0.5,act by +38 mV (Fig. 4C), the same shift as that observed with WT hERG channels. Cd2+ also slowed the rate of current activation (Fig. 4A and D) and accelerated the rate of current deactivation (Fig. 4E) at all potentials examined. For example, the fast and slow time constants (fast and slow, respectively) for deactivation at –110 mV were 51 ± 8 ms and 314 ± 7 ms in control, compared to 16 ± 3 and 130 ± 16 ms after Cd2+ (n = 3), respectively. At –80 mV, fast was 212 ± 58 ms and slow was 656 ± 166 ms in control, compared to a fast value of 38 ± 6 ms and a slow value of 163 ± 14 ms after application of Cd2+ (n = 3). Thus, Cd2+-induced changes in deactivation and shifts in the voltage dependence of activation gating of inactivation-deficient G628C/S631C channels were the same as WT hERG channels.
, 百拇医药
A, currents conducted by G628C/S631C channels before (left) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (right). Currents were elicited with 2-s pulses to potentials from –50 mV to +50 mV, applied in 20-mV increments. Tail currents were measured at –120 mV. B, normalized I–V relationships for peak currents elicited with 2-s pulses before () and after application of 0.5 mM Cd2+ (). Currents were normalized relative to the peak current at +50 mV in the absence of Cd2+. C, g–V relationship before () and after application of 0.5 mM Cd2+ (). Control: V0.5,act, –27.5 ± 0.6 mV; k, 11.0 ± 0.3 mV; after application of Cd2+: V0.5,act, +10.4 ± 0.6 mV; k, 13.9 ± 2.5 mV (n = 4). D, time constants for activation of G628C/S631C hERG channel currents before (open symbols) and after application of Cd2+ (filled symbols). E, tail currents recorded at potentials of –120 to –30 mV, applied in 10-mV increments. Top, voltage pulse protocol; middle, control currents; bottom, currents in presence of 0.5 mM Cd2+.
, 百拇医药
Site-directed mutagenesis of potential Cd2+-binding residues in the extracellular linkers of the hERG channel
The coordination geometry of Cd2+ in metalloproteins is either tetrahedral or octahedral and the most frequent residues involved in Cd2+ coordination are Cys, His, Asp and Glu (Rulisek & Vondrasek, 1998). Such residues that were accessible from the extracellular solution and could potentially interact with extracellular Cd2+ were mutated to Ala. Residues located on the N-terminus, C-terminus or intracellular linkers were assumed to be inaccessible to extracellular Cd2+ during the short timecourse of the experiments and therefore, were not investigated. The approximate locations of the Ala-substituted residues, including five Cys, three His, nine Asp and eight Glu are illustrated in Fig. 1B. Residues that when mutated to Ala reduced the 0.5 mM Cd2+-induced shift in the voltage dependence of hERG channel activation were considered to be potential coordination sites for Cd2+. A Cd2+ concentration of 0.5 mM was used because it was high enough to cause a large shift in V0.5,act (+36 mV) of WT channels, yet low enough to minimize non-specific binding and screening of negative surface charges. Mutation of a critical Cd2+-binding residue would be expected to reduce binding affinity. However, because 0.5 mM is near the EC50 for the effects of Cd2+ on WT hERG channels, a mutation-induced change in the shift in V0.5,act could reflect a change in binding affinity, maximum shift or a combination of the two effects.
, http://www.100md.com
Fourteen candidate Cd2+-binding residues are located in the extracellular linkers of hERG channel (Fig. 1B). Single or combined mutations were used to assess the potential role of these residues in Cd2+ binding. Four Glu and two Cys residues are located in the S1–S2 linker. Instead of assaying six different channels containing single mutations, we constructed a mutant channel with all six residues mutated to Ala (‘4E2C/6A’ hERG). This mutant channel conducted currents that were similar to WT hERG channels. It is surprising that neutralization of 16 negative charges (4 Glu x 4 subunits) near the extracellular surface of each channel did not alter V0.5,act or the response to 0.5 mM Cd2+. The Cd2+-induced shifts in the isochronal (Fig. 5B) and fully activated I–V relationships (Fig. 5C), V0.5,act (Table 1) and deactivation kinetics (Fig. 5D) of 4E2C/6A hERG were the same as in WT hERG channels. The 4E2C/6A mutations did not alter the Cd2+-induced shift in V0.5,inact (Table 2). Thus, mutation of all the residues in the S1–S2 linker to Ala did not alter the effects of Cd2+ on channel gating.
, 百拇医药
A, currents conducted by 4E2C/6A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from –70 mV to +30 mV, applied in 20-mV increments. B–D, effect of 0.5 mM Cd2+ on 4E2C/6A currents: isochronal I–V relationships normalized to peak current (Ipeak) under control conditions (B); fully activated I–V relationships normalized to the peak tail current (Itail) measured under control conditions (C); and time constants of current deactivation for 4E2C/6A currents before and after application of Cd2+. Left-hand axis indicates values for slow and right-hand axis is for fast (D). In each panel, open symbols represent data for control currents and filled symbols represent measurements in the presence of 0.5 mM Cd2+. E–H, the same protocols and analyses for E518A hERG channel currents. Current traces in E correspond to test potentials of –60 to +60 mV in 20-mV increments for both conditions, before (top) and after adding 0.5 mM Cd2+ (bottom). I–P, the same protocols and analyses for D580A (I–L) and H587A (M–P) hERG channel currents. Current traces for D580A (I) and H587A (M) mutant channels correspond to test potentials of –50 to +50 mV applied in 20-mV increments. Note the lack of effect of 0.5 mM Cd2+ on fully activated I–V relationship for D580A mutant channels (K) that results from a reduced effect on inactivation (Table 2).
, 百拇医药
Two candidate Cd2+-binding residues (E518 and E519) are located in the S3–S4 linker. Mutation of these residues were previously reported to alter the extracellular Ca2+-induced changes in hERG channel gating (Johnson et al. 2001). We found that the biophysical properties of E519A hERG channels were similar to WT, whereas the E518A mutation shifted the peak of the I–V relationship (Fig. 5F) and V0.5,act by +26 mV (Table 1) and accelerated the rate of deactivation (Fig. 5H). Despite differences in biophysical properties, E518A (Fig. 5E–H) and E519A (not shown) mutant channels retained a normal 0.5 mM Cd2+-induced effect on activation (Table 1) and inactivation (Table 2).
, 百拇医药
There are six potential Cd2+-interacting residues located in the S5–P region, including E575, H578, D580, H587, D591 and D609 (Fig. 1B). A previous study reported that mutation of some of these residues to Cys (i.e. E575, H578, D580 and D609) did not affect inactivation, whereas D591C and H587C disrupted inactivation (Liu et al. 2002). Mutation of these residues to Ala did not significantly alter the baseline biophysical properties of the channels, or the effect of 0.5 mM Cd2+ on V0.5,act compared to WT the hERG channel (Table 1). For example, current traces, I–V relationships and kinetics of deactivation are shown in Fig. 5I–L for D580A channels and in Fig. 5M–P for H587A channels. The effects of Cd2+ on H587A channel properties were normal. However, Cd2+ reduced D580A tail-current amplitudes by 16% (Fig. 5I) and did not cause a positive shift in the voltage dependence of the fully activated I–V relationship (Fig. 5K). E575A and H578A channel currents were also reduced by Cd2+. The V0.5,inact for E575A, H578A and D580A was only shifted by +7 to +11 mV in response to 0.5 mM Cd2+ (Table 2). Combining two of the mutations (H578A/D580A) or all three mutations (E575/H578/D580) into a single channel did not have an additive effect. The Cd2+-induced shift in V0.5,inact for the double and triple mutants was +8 mV, similar to the shift caused by single mutations (Table 2).
, 百拇医药
In summary, mutations of the 14 residues located in the extracellular linkers of hERG channel did not significantly alter Vact induced by Cd2+. However, mutation of E575, H578 or D580 reduced the Cd2+-induced shift in V0.5,inact, confirming the important role of the S5-P linker in rapid C-type inactivation of hERG channels (Liu et al. 2002).
Site-directed mutagenesis of potential Cd2+-binding residues in the transmembrane domains of the hERG channel
, 百拇医药
The only potential Cd2+-interacting residue located in S1 is D411 (Fig. 1B). Mutation of this residue to Ala caused a –17.5 mV shift in V0.5,act (Table 1), a +15 mV shift in V0.5,inact (Table 2), an enhanced rate of activation and deactivation (Fig. 6A and D), and a slight leftward displacement of the isochronal and fully activated I–V relationship (Fig. 6B and C). The Vact for D411A channels was identical to the value for WT channels (+36 mV), but the Vinact was 10 mV greater for the mutant channel (+28 mV) than for the WT channel (+18 mV). There are three potential Cd2+-interacting residues located in S5: C555, H562 and C566 (Fig. 1B). No hERG channel currents were detected in oocytes expressing H562A channels. The biophysical properties of C566A channel currents and their response to 0.5 mM Cd2+ were similar to WT channels (Fig. 6E–H and Tables 1 and 2). C555A (not shown) was also similar to WT. Two potential Cd2+-interacting residues (E637 and C643) are located in S6 (Fig. 1B). E637A channels did not functionally express, but C643A channels had normal biophysical properties and responsiveness to Cd2+ (Fig. 6I–L and Tables 1 and 2). Thus, with the exception of H562A and E637A, which did not functionally express, mutation of Asp or Cys residues located in the S1, S5 and S6 did not diminish the changes in gating caused by Cd2+.
, 百拇医药
A, currents conducted by D411A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from –60 to +40 mV, applied in 20-mV increments. B–D, effect of 0.5 mM Cd2+ on D411A currents: isochronal I–V relationships normalized to peak current under control conditions (B); fully activated I–V relationships normalized to the peak current measured under control conditions (C); and time constants for fast (triangles) and slow (inverted triangles) components of current deactivation for D411A currents. Left-hand axis indicates values for slow and right-hand axis is for fast (D). In each panel, open symbols represent data for control currents and filled symbols represent measurements in the presence of 0.5 mM Cd2+. E–H, the same protocols and analyses for C566A hERG channel currents. Current traces in E correspond to test potentials ranging from –50 to +50 mV applied in 20-mV increments. I–L, the same protocols and analyses for C643A hERG channel currents. Test potentials used for C643A mutant channel (I) are the same as those used for D411A mutant described in A.
, 百拇医药
There are three potential Cd2+-interacting residues located in S2, including D456, D460 and D466 (Fig. 1B). As expected from previous studies of divalent cation interactions with Shaker (Papazian et al. 1995; Tiwari-Woodruff et al. 2000) and dEAG channels (Silverman et al. 2000, 2004), mutation of these Asp residues in hERG channel caused significant changes in channel gating. D456A and D460A channels deactivated more rapidly than normal and the I–V relationships were shifted towards the right compared to the control(Fig. 7A–H). The V0.5,act was shifted by +34 mV for D456A and +22 mV for D460A (Table 1). As in dEAG, mutation of these Asp residues reduced the effects of extracellular Cd2+. The Vact induced by Cd2+ was reduced to +16 mV for D456A and +19 mV for D460A (Table 1). The reduced effect of Cd2+ on Vact together with a normal Vinact (Table 2) caused the peak of the relative I–V relationship in the presence of Cd2+ to be larger compared to other mutant channels (Fig. 7B and F). Moreover, the Cd2+-induced shift in the voltage dependence of deactivation kinetics for these two mutants (Fig. 7D and H) was reduced compared to Vact. Of the three Asp residues in S2, D466 is positioned closest to the C-terminal end of the domain and would be expected to be the least accessible to extracellular divalent cations. Although D466A channels activated and deactivated rapidly (Fig. 7I and L), the V0.5,act was normal (–24 mV) as was the Cd2+-induced shift in the voltage dependence of activation (Vact, +38 mV; Table 1) and deactivation kinetics (Fig. 7L). By contrast, the V0.5,inact of D466A channels was +12 mV more positive than normal, and Cd2+ induced a larger than normal Vinact (+ 28 mV, Table 2). Thus, mutation of the two outermost Asp residues in S2 (D456 and D460) reduced the effects of Cd2+ on hERG channel activation.
, 百拇医药
A, currents conducted by D456A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from –30 to +70 mV, applied in 20-mV increments. B–D, effect of 0.5 mM Cd2+ on D456A currents: isochronal I–V relationships normalized to peak current under control conditions (B); fully activated I–V relationships normalized to the peak current measured under control conditions (C); and time constants for fast (triangles) and slow (inverted triangles) components of current deactivation for D456A currents. Left-hand axis indicates values for slow and right-hand axis is for fast (D). In each panel, open symbols represent data for control currents and filled symbols represent measurements in the presence of 0.5 mM Cd2+. E–H, the same protocols and analyses for D460A hERG channel currents. I–P, the same protocols and analyses for D466A (I–L) and D509A (M–P) hERG channel currents. Examples of current traces for D460A, D466A and D509A mutant channels (E, I and M) correspond to test potentials of –50 to +50 mV applied in 20-mV increments.
, 百拇医药
Two potential Cd2+-binding residues, D501 and D509, are located in the S3 domain (Fig. 1B). D501 channels expressed poorly, but in oocytes with very small endogenous currents we determined that V0.5,act was 42 mV more positive than in WT channels. Despite such a positive V0.5,act, the response to 0.5 mM Cd2+ was equivalent to the response in WT hERG channels, including a +37 mV shift in V0.5,act (Table 1). D509A channels activated at potentials 27 mV more positive than in WT channels and were the least sensitive of all the mutant channels to the effects of Cd2+ on activation gating (Fig. 7M–P). V0.5,act was only shifted +8.1 mV by 0.5 mM Cd2+ (Table 1). By contrast, the V0.5,inact and the Cd2+-induced Vinact values were normal for D509A channels (Table 2). The differential effects of Cd2+ on activation and inactivation gating of D509A channels resulted in an increased peak in the I–V relationship. Mutation of both Asp residues in S3 affected channel gating; however, only mutation of the residue located nearest the extracellular space (D509) diminished the functional response to Cd2+.
, http://www.100md.com
In the majority of metalloproteins, Cd2+ usually binds to Cys residues with equal or greater affinity than to Asp residues (Rulisek & Vondrasek, 1998). Therefore, we individually mutated D456, D460 and D509 to Cys in an attempt to increase the Vact induced by Cd2+. For WT hERG channels, 30 μM and 100 μM Cd2+ induced a Vact of 11.8 ± 0.8 mV (n = 10) and 21.1 ± 0.6 mV (n = 7), respectively. The sensitivity of D460C to Cd2+ was markedly enhanced. For this mutant channel, Vact was increased to 29.3 ± 1.7 mV for 30 μM Cd2+ and 61.3 ± 0.8 mV for 100 μM Cd2+ (n = 5 for each concentration). By contrast, D456C and D509C hERG channels were less responsive than WT channels. The Vact in response to 100 μM Cd2+ was 14.8 ± 1.1 mV for D456C hERG (n = 5) and 13.4 ± 0.5 mV for D509C hERG channels (n = 4). Thus, the sensitivity of hERG channel activation to Cd2+ can be increased by mutation of D460, but not D456 or D509 to Cys.
, 百拇医药
In summary, based on the reduced sensitivity of mutant channels to externally applied Cd2+, two acidic residues in S2 (D456 and D460) and one acidic residue in S3 (D509) contribute to a binding site that mediates the effects of Cd2+ on activation gating of hERG channels. We next determined how mutation of two or all three of these residues affected the response to Cd2+.
Combined mutation of acidic residues in S2 and S3
Three double mutant channels (D456A/D460A, D456A/D509A and D460A/D509A) were constructed. The V0.5,act of D456A/D509A channels was +56 mV more positive than of WT channels and nearly equivalent to the sum of the Vact values induced by the individual mutations (+59 mV). The V0.5,act values for D456A/D460A (+62 mV) and D460A/D509A (+47 mV) channels were shifted rightwards more than expected from the sum of the single mutations. The rate of deactivation was markedly enhanced for all three double mutant channels (Fig. 8A, C and E), similar to the effect caused by mutations of the single residues (Fig. 7). The Vact induced by 0.5 mM Cd2+ was +8.4 mV for D460A/D509A, +4.1 mV for D456A/D509A and +2.5 mV for D456A/D460A channels (Table 1). D460A/D509A channels retained the same sensitivity to Cd2+ as the D509A channels (Vact of +8 mV for both), suggesting that D460 is the least important of the three residues for coordination of Cd2+. At the very positive potentials required to activate the double mutant channels, contamination by large endogenous outward currents prevented an accurate measurement of V0.5,inact. However, 0.5 mM Cd2+ increased the amplitude of tail currents (Fig. 8B, D and F). This finding was consistent with a positive shift in the voltage dependence of inactivation in the absence of a significant effect on activation. Combined mutation of all three residues (D456A/D460A/D509A) further stabilized the closed state of the channel which could only be activated with very strong depolarizations. At these potentials, the large outward currents reflect activation of endogenous channels; however, the endogenous channels deactivate very rapidly compared to hERG (Fig. 8G) and do not significantly contaminate the tail currents used to measure the voltage dependence of activation. The estimated V0.5,act for the triple mutant channel was +125 mV and the 0.5 mM Cd2+-induced Vact was < 3 mV (Fig. 8H). Together, these findings suggest a crucial role for D456, D460 and D509 in the normal gating of hERG channels and that Cd2+ ions interact with all three residues in the WT channel.
, 百拇医药
A, currents conducted by D456A/D460A channels before (top) and after equilibration of oocyte with solution containing 0.5 mM Cd2+ (bottom). Currents were elicited with 2-s pulses to potentials from +30 to +110 mV, applied in 20-mV increments. Tail currents were measured at –70 mV. B, Cd2+ enhanced the tail current amplitude of D456A/D460A channels, but did not appreciably shift V0.5,act. C–F, the same protocols and analyses for D456A/D509A (C and D), D460A/D509A (E and F) and D456A/D460A/D509A (G and H) hERG channel currents. Examples of current traces for D456A/D509A and D460A/D509A (C and E) correspond to test potentials from 0 to +80 mV applied in 20-mV increments. D456A/D460A/D509A currents (G) were recorded in 20-mV increments to test potentials of +70 to +150 mV. Tail currents were measured at –20 mV.
, 百拇医药
Concentration-dependent effects of Cd2+ on hERG channel activation
Neutralization of D456 and D460 reduced Vact from +36 mV to +2.5 mV, suggesting that the high-affinity binding site for Cd2+ was effectively eliminated by this double mutation. However, these mutations were not expected to prevent higher concentrations of Cd2+ or other divalent cations from screening non-specific negative surface charges that exist near the extracellular regions of the channel. To differentiate between the gating effects caused by high-affinity binding to the putative S2–S3 pocket and screening of non-specific negative surface charge, the Vact induced by a range of [Cd2+]e(10 μM–30 mM) was determined for WT channels (Fig. 9A, ) and compared to D456A/D460A hERG channels (Fig. 9A, ). The Vact induced by 30 mM Cd2+ was +80 mV for WT channels and +23 mV for D456A/D460A channels, but even this high cation concentration did not cause a saturating effect. For the purpose of analysis, we assumed that the Vact induced by Cd2+ for D456A/D460A channels was caused by charge screening of non-specific negative surface charges and that a plot of the difference between the Vact–[Cd2+]e relationship for WT and D456A/D460A channels (Vact,subtract) versus [Cd2+]e was an estimate of the concentration–response relationship for Cd2+ binding to the S2–S3 site (Fig. 9B). The value of Vact,subtract saturated at 10 mM Cd2+. When fitted with a Hill equation, the maximum Vact,subtract was +58 mV, the EC50 for Cd2+ was 345 μM and the Hill coefficient was 0.59. The same subtraction and fitting procedure was repeated for the other mutant channels (Fig. 9C–F). Most mutations did not alter the concentration–response to Cd2+ (Table 3). As examples, the analyses for E518A and D591A mutant channels are shown in Fig. 9C. As expected, the D456A, D460A and D509A mutations reduced the efficacy (maximum response) and potency (EC50) of the effects of Cd2+ on channel activation (Fig. 9D). D509A had the greatest effect of the three mutations, reducing the maximum Vact,subtract to +16 mV and increasing the EC50 to 1.34 mM (Table 3). The double mutation of D460A/D509A was similar (data not shown), with a maximum Vact of +17 mV and an EC50 of 1.04 mM (Table 3).
, 百拇医药
A, Vact for WT hERG channels () and D456A/D460A channels (). Cd2+-induced Vact in this double mutant was considered to come from screening of non-specific surface charges. B, by subtracting Vact for D456A/D460A from Vact of WT channels we obtained an estimate of Cd2+-induced Vact (Vact,subtract) that results from specific binding to acidic residues in S2 and S3. The same procedure was applied to obtain the data presented in panels C–F. C, Vact,subtract for WT (), E518A () and D591A () channels. D, Vact,subtract for WT (), D456A (), D460A () and D509A () channels. E, Vact,subtractfor WT (), D411A () and D466A () channels. F, Vact,subtract for WT (), E519A () and C566A () channels. n = 4–13 for each concentration of Cd2+.
, http://www.100md.com
Unexpectedly, Vact,subtract for D411A and D466A channels was greater than WT channels when [Cd2+]e > 3 mM (Fig. 9E). These mutated residues are located near the cytosolic ends of the S1 and S2 domains in the hERG channel. The maximum Vact,subtract was +78 mV for D411A and +81 mV for D466A channels, an increase of about 20 mV relative to WT channels. In addition, both mutations caused a 2.75-fold decrease in Cd2+ potency (Table 3). Finally, two mutations (E519A and C566A) induced a small increase in Cd2+ potency without a significant change in maximum response (Fig. 9F and Table 3).
, 百拇医药
Ca2+ and Mg2+ interaction sites
The effects of 10 mM extracellular Ca2+ and 20 mM extracellular Mg2+ on Vact of WT and selected mutant hERG channels were determined. These concentrations are about 10-times greater than normal physiological concentrations. The effects of Ca2+ and Mg2+ were determined on the mutant channels that were most affected by Cd2+, including D411A, D456A, D460A, D466A, D509A, E518A and E519A. The Vact for WT channels was 26.1 mV for 10 mM Ca2+ (Table 4), a concentration about 100-times greater that the [Cd2+]e required for an equivalent shift in V0.5,act. The Ca2+-induced Vact was altered by mutation of D456, D460 and D466 in S2, D509 in S3, and E518A in the S3–S4 linker. The Vact for single and double mutants can be compared to estimate the relative importance of D456, D460 and D509 for binding of Ca2+. D456A/D460A channels were slightly less sensitive to Ca2+ than channels with either single mutation. D460A/D509A channels had the same Ca2+ sensitivity as mutation of D509 alone, whereas the combined effects of D456A and D509A were additive. Taken together these findings indicate that D456 and D509 are the most important residues with regard to Ca2+-induced alteration of hERG channel activation.
, 百拇医药
Mg2+ was less effective than Ca2+ in affecting hERG channel activation. A concentration of 20 mM Mg2+ was required for a Vact of +14.0 mV for WT channels. Unlike Ca2+, channel sensitivity to Mg2+ was not affected by D509A, and was enhanced by E518A (Table 4). D456A was the only mutation that significantly decreased the effects of Mg2+ on hERG channel gating, causing a reduction of Vact to 6.3 mV (Table 4). Double mutations of Asp residues in the S2 and S3 domains, especially D456/D460 and D456/D509, caused a synergistic decrease in the sensitivity to Mg2+. Although the functional consequences of single or double mutations differ for Cd2+, Ca2+ and Mg2+, our findings suggest that all three cations can interact with a coordination site formed by D456 and D460 residues in the S2 domain and D509 in the S3 domain.
, 百拇医药
Discussion
Effects of Cd2+ and other divalent cations on ERG channels
Extracellular divalent cations have multiple effects on voltage-gated ion channels. These cations can directly block the pore (Armstrong et al. 1982), screen negative surface charge and alter the apparent transmembrane voltage detected by intramembrane charged residues in the voltage sensors (McLaughlin, 1989), or bind to specific sites on the channel protein and alter the voltage dependence and kinetics of gating (Gilly & Armstrong, 1982; Armstrong & Lopez-Barneo, 1987; Begenisich, 1988). In this study we used site-directed mutagenesis to define the potential binding sites on the hERG channel that mediate the Cd2+-induced alterations in activation and inactivation gating. The obvious candidate residues involved with charge screening are acidic amino acids located in the extracellular linkers between the six transmembrane domains of each subunit. It is surprising that individual mutations of the four acidic amino acids in the S5–P linker to Ala did not significantly alter the ability of Cd2+ to shift V0.5,act. This finding is in contrast to the situation in Kv1–4 channels where the first five residues in the S5-P linker were proposed to be the main structural determinants of divalent cation-induced shifts of V0.5,act (Elinder et al. 1996, 1998). It was also surprising that combined neutralization of four negative charges in the S1–S2 linker did not cause a shift in the voltage dependence of hERG channel gating. Thus, the effects of relatively low concentrations (0.5 mM) of Cd2+ do not involve the 16 acidic residues located in the S1-S2 or the 16 acidic residues in the S5–P linkers present in a homotetrameric hERG channel.
, 百拇医药
Cd2+ slowed the activation, increased the rate of deactivation and induced positive shifts in the voltage dependence of activation and inactivation of hERG channel. In cardiac myocytes, ERG channels conduct the rapid delayed rectifier K+ current, IKr (Sanguinetti et al. 1995; Trudeau et al. 1995). The effects of Cd2+ on IKr was first studied by Follmer et al. (1992) who reported that 0.2 mM Cd2+ accelerated deactivation, shifted V0.5,act to more positive potentials and enhanced tail current magnitude in cat ventricular myocytes. The increase in current was attributed to a reduction in channel rectification. Similar effects of Cd2+ were reported for IKr recorded from guinea-pig myocytes (Daleau et al. 1997) and of Cd2+, Ni2+, Mn2+ and Co2+ in rabbit ventricular myocytes (Paquette et al. 1998). Co2+ was initially reported to block IKr (Sanguinetti & Jurkiewicz, 1991), but later experiments with rabbit IKr (Paquette et al. 1998) and hERG channels expressed in oocytes (Sanchez-Chapula & Sanguinetti, 2000) demonstrated that the decrease in current magnitude was caused by a positive shift in the voltage dependence of channel activation. Ca2+, Mg2+, Ba2+ and Sr2+ have similar, but much weaker effects than Co2+ or Cd2+ on IKr (Paquette et al. 1998). Although the effects of Ca2+ and Mg2+ on IKr or hERG channels was misinterpreted as a time- and voltage-dependent channel block that defines channel gating under normal physiological conditions (Ho et al. 1996, 1998; Song et al. 1999), some divalent cations can block hERG. For example, Zn2+ can block IKr in addition to modulating channel gating (Paquette et al. 1998).
, 百拇医药
Johnson et al. (1999) reported that a [Cd2+]e 0.2 mM caused a positive shift in the voltage dependence of rapid inactivation without affecting activation and that an inactivation-removed mutant hERG channel did not respond to 0.2 mM Cd2+. These findings were the basis of a model where low concentrations of Cd2+ had a direct effect on inactivation and that a lower affinity binding site and/or non-specific charge screening mediated the observed effects on activation gating. By contrast, we found that 0.5 mM Cd2+ slowed activation, accelerated deactivation and caused a +38 mV shift in V0.5,act of G628C/S631C hERG, effects that were similar to WT hERG channels. Nonetheless, as discussed below, some of our findings also support the idea that Cd2+ binds to two separate sites on the hERG channel.
, http://www.100md.com
Binding of Cd2+ to acidic residues in the S2 and S3 domains mediates the positive shift in hERG channel activation
External divalent cations can bind to specific sites on Kv channels to affect gating. This has been best studied in Shaker and dEAG channels by Papazian and colleagues (Tang et al. 2000; Silverman et al. 2000, 2004). In Shaker channels, acidic residues in S2 and S3 (E283, E293 and D316) are required for normal activation and channel folding and trafficking to the cell membrane (Tiwari-Woodruff et al. 1997). These effects are mediated by electrostatic interactions between these S2–S3 acidic residues and basic residues in S4 (Tiwari-Woodruff et al. 2000). In dEAG channels, neutralization of D278 in S2 and D327 in S3 slows channel activation and reduces or abolishes the slowing of activation caused by Mg2+ (Silverman et al. 2000, 2003, 2004). These findings suggest that Mg2+ (and Mn2+ and Ni2+) shield the negative side chains of D278 and D327 and prevent charge pairing with specific basic residues in the S4 domain. D278E abolishes Mg2+ but not Mn2+ sensitivity of EAG channels. The reason for this difference is unclear, but might be related to the ability of Mn2+ but not Mg2+ to be coordinated by a distorted octahedral geometry (Silverman et al. 2004). Coordination of Mg2+ by D278 and D327 in dEAG implies that these residues are positioned within a few angstroms of each other, constraining the structural arrangement of S2 and S3. Together with other findings, these constraints were used to construct a generalized structural model of the S1–S4 domains for Kv channels (Silverman et al. 2004) that differs markedly from the structure of the voltage sensor deduced by X-ray crystallography of the KvAP bacterial K+ channel (Jiang et al. 2003a,b).
, 百拇医药
Cd2+ binds to proteins in either a tetrahedral or octahedral arrangement and binding is most often mediated by interaction with the S of Cys, the N or N in the imidazole ring of His, or bidentate interactions with the O and N of the side chains of single Asp or Glu residues (Rulisek & Vondrasek, 1998). To identify the binding site for extracellular Cd2+ on the hERG channel, we mutated to Ala all potential coordination residues, including five Cys, three His, nine Asp and eight Glu residues located in transmembrane domains or extracellular loops linking these domains. Some of these mutations altered the normal gating of hERG channels and diminished the response to Cd2+.
, 百拇医药
Mutation of specific residues located in S2 (D456 and D460) or S3 (D509) domains greatly reduced the Cd2+-induced shift in V0.5,act. For example, the V0.5,act for D509A channels was shifted by +26 mV compared to the value in WT channels, but 0.5 mM Cd2+ only caused a further +8 mV shift. It could be argued that the ability of Cd2+ to shift V0.5,act was somehow impaired by mutations that altered the voltage dependence of activation. Evidence against this possibility is provided by two mutant channels. First, E518A channels also had a positive value for V0.5,act compared to WT channels, but responded to 0.5 mM Cd2+ with a +40 mV shift in V0.5,act, equivalent to that measured for WT channels. Second, the V0.5,act for D411A channels was 20 mV more negative than for WT channels, but 0.5 mM Cd2+ induced a normal +36-mV shift in this parameter. Thus, the response to Cd2+ was not correlated with the baseline value of V0.5,act. The site-directed mutagenesis approach is an indirect method to define residues important for ligand binding. An obvious limitation of the method is that it cannot distinguish between direct and allosteric effects. However, the specific mutations that most altered the response to Cd2+ were of residues that cluster together in relatively close proximity. This finding strongly suggests that D456, D460 and D509 form a binding site for Cd2+.
, 百拇医药
We were fortunate that most mutant channels functionally expressed in oocytes and retained relatively normal gating. This is in contrast to an earlier study where hERG channel residues were mutated to Cys and it was found that D456C was non-functional. Furthermore, because substitution of other S2 and S3 Asp residues with Cys did not alter the Ca2+-induced shift in V0.5,act of hERG, it was concluded that these Asp residues did not form a metal ion-binding site (Liu et al. 2003). Because Cys might adequately substitute for one of the Asp residues in coordination of Ca2+, Cys-substitution may not be a valid test of the ability of a specific residue to cooperate in divalent cation binding. In our study, D456 was mutated to Ala. D456A channels expressed well in oocytes and this mutation interfered with the Cd2+-induced shift in V0.5,act, as did mutations of other S2 and S3 acidic residues.
, http://www.100md.com
Combined mutation of multiple acidic residues provided further evidence for a Cd2+ coordination site in a pocket formed by the S2 and S3 domains. The shift in V0.5,act for D456A/D509A channels was equivalent to the sum of the shifts induced by individual mutations, whereas D456A/D460A and D460A/D509A channels were shifted more than expected from the sum of the shift of the single mutations. This finding suggests that the functional effects of mutations of D456 or D509 are independent and that efficient coordination of V0.5,act Cd2+ in the putative pocket formed between S2 and S3 is most dependent on these two residues. The Cd2+-induced shift in activation gating that remained after mutation of both D456 and D460 probably results from screening of sialic acid, the typical terminal residue on cell surface oligosaccharides and other negatively charged molecules bound to the extracellular cell surface.
, 百拇医药
Charge pairing between acidic residues in S2–S3 and basic residues in S4 can serve to promote or stabilize voltage sensor movement and/or facilitate correct protein folding and trafficking of the channel complex to the cell surface. Our finding that the triple mutant (D456A/D460A/D509A) functionally expressed indicates that charge pairing between basic residues of S4 and these acidic residues is not an absolute requirement for protein folding, trafficking or gating.
, http://www.100md.com
Cd2+ is commonly coordinated in a tetrahedral arrangement in metalloproteins (Rulisek & Vondrasek, 1998). It is possible that extracellular Cd2+ may also be coordinated in a tetrahedral arrangement in a pocket between S2 and S3 domains of each hERG subunit and that three of the coordination residues are D456, D460 and D509. The identity of the putative fourth coordination point is unknown, but could be a water molecule or a peptidyl oxygen of another unidentified residue. Double mutations of D456, D460 or D509 suggest that these three residues also contribute to a lower-affinity binding site for Ca2+ and Mg2+ and that D460 is the least important of the three residues for coordinating divalent cations. It seems likely that all divalent cations that affect activation gating of hERG channel, including Ca2+, Mg2+, Ba2+, Sr2+, Mn2+, Co2+, Cd2+, Zn2+, and Ni2+ (Follmer et al. 1992; Paquette et al. 1998; Ho et al. 1998, 1999; Johnson et al. 1999, 2001; Sanchez-Chapula & Sanguinetti, 2000), bind to the same S2/S3 site.
, 百拇医药
Coupling between activation and inactivation
For some residues, the mutation-induced shift in V0.5,act and V0.5,inact were the same. For example, D456A shifted V0.5,act by +33 mV and V0.5,inact by +29 mV. For some residues, Ala substitution had a differential effect on the voltage dependence of activation and inactivation. For example, D411A shifted V0.5,act by –19 mV, but shifted V0.5,inact by +14 mV. D460A and D509A shifted V0.5,act by +22 and +27 mV, but had no significant effect on V0.5,inact. We have previously noted a disconnection between the effects of a point mutation on activation and inactivation gating. S631A shifts V0.5,inact by +100 mV but has no effect on V0.5,act (Zou et al. 1998) and R531A shifts V0.5,act by +57 mV without affecting V0.5,inact (Piper et al. 2005). As shown in the present study, extracellular cations can also differentially affect gating associated with activation and inactivation. For example, 0.5 mM Cd2+ shifted V0.5,act of WT hERG channel by +36 mV, twice as much as the shift in V0.5,inact (+18 mV). Thus, shifts in the voltage dependence of inactivation induced by mutation or divalent cations can differ significantly from the voltage dependence of activation. This difference could be caused by alteration of the coupling between activation and inactivation and does not necessarily imply that the two gating processes are controlled by different voltage sensors (Piper et al. 2003).
, 百拇医药
The binding site for the Cd2+-mediated shift in hERG channel inactivation remains elusive
Cd2+ destabilizes the inactivated state of the hERG channel. A poorly defined rearrangement of the S5-P linker/pore helix is presumed to be responsible for rapid C-type inactivation (Smith et al. 1996; Herzberg et al. 1998; Ficker et al. 2001; Liu et al. 2002). The S5-P linker of hERG channel is 42 amino acids in length and considerably longer than most other Kv channels that have a S5-P linker length of < 15 residues. [2-(trimetylammonium) ethyl] methane sulfonate (MTSET) modification of Cys residues introduced into the middle of the hERG linker (residues 583–597) disrupted channel inactivation and this region was proposed to form an -helix that bridged between the voltage sensor (S4) and the outer mouth of the channel pore (Liu et al. 2002). In contrast, modification of residues closer to S5 (residues 574–582) by MTSET had only small effects on gating and were proposed to face away from the channel and towards the extracellular aqueous solution. The structure of the isolated S5-P linker, determined by NMR spectroscopy, was disordered in an aqueous solution (Torres et al. 2003). However, in the presence of micelles, the linker formed two helical regions (Trp585–Ile593 and Gly604–Tyr611). The first helix was preceded by a highly flexible stretch of amino acids (from Ala571 to Gly585) (Torres et al. 2003) that has previously been reported to be little affected by thiol modification of introduced Cys residues (Liu et al. 2002). Mutation of E575, H578 or D580 to Ala partially inhibited the effects of Cd2+ on inactivation. Based on NMR findings, these three residues are located in a disordered region of the S5-P linker that is readily accessible by extracellular Cd2+.
, http://www.100md.com
Two findings suggest that the effect of Cd2+ on inactivation is mediated by binding to a site distinct from the S2–S3 pocket. First, mutation analysis of D456, D460 and D509 indicates that Cd2+ can shift V0.5,inact independently of its effects on activation gating. Second, mutation of E575, H578 or D580 reduced the Cd2+-induced shift in V0.5,inact, but did not alter the Cd2+-induced shift in V0.5,act. Ala substitution of these three residues did not alter V0.5,inact in the absence of Cd2+, suggesting that these specific residues are not crucial for normal coupling between activation and inactivation. However, combined mutations of all three residues did not prevent the Cd2+-induced shift in V0.5,inact as would be predicted if these residues formed a binding pocket. Cd2+ may bind to another unidentified site to affect inactivation gating.
, 百拇医药
Summary
Mutation of three acidic residues in S2 and S3 of the hERG channel (D456, D460 and D509) greatly attenuated the extracellular Cd2+-induced positive shifts in the voltage dependence of activation but not inactivation. We conclude that coordination of Cd2+ by these acidic residues prevent specific charge pair interactions with basic residues in S4 that normally facilitates hERG channel gating associated with channel activation.
, http://www.100md.com
References
Anumonwo JM, Horta J, Delmar M, Taffet SM & Jalife J (1999). Proton and zinc effects on HERG currents. Biophys J 77, 282–298.
Armstrong CM & J (1987). External calcium ions are required for potassium channel gating in squid neurons. Science 236, 712–714.
Armstrong CM, Swenson RP Jr & Taylor SR (1982). Block of squid axon K channels by internally and externally applied barium ions. J Gen Physiol 80, 663–682.
, 百拇医药
Begenisich T (1988). The role of divalent cations in potassium channels. Trends Neurosci 11, 270–273.
Cherny VV & DeCoursey TE (1999). pH-dependent inhibition of voltage-gated H+ currents in rat alveolar epithelial cells by Zn2+ and other divalent cations. J Gen Physiol 114, 819–838.
Daleau P, Khalifa M & Turgeon J (1997). Effects of cadmium and nisoldipine on the delayed rectifier potassim current in guinea pig ventricular myocytes. J Pharmacol Exp Ther 281, 826–833.
, 百拇医药
Elinder F, Liu Y & Arhem P (1998). Divalent cation effects on the Shaker K channel suggest a pentapeptide sequence as determinant of functional surface charge density. J Membr Biol 165, 183–189.
Elinder F, Madeja M & Arhem P (1996). Surface charges of K channels. Effects of strontium on five cloned channels expressed in Xenopus oocytes. J Gen Physiol 108, 325–332.
Ficker E, Jarolimek W & Brown AM (2001). Molecular determinants of inactivation and dofetilide block in ether a-go-go (EAG) channels and EAG-related K+ channels. Mol Pharmacol 60, 1343–1348.
, 百拇医药
Follmer CH, Lodge NJ, Cullinan CA & Colatsky TJ (1992). Modulation of the delayed rectifier, IK, by cadmium in cat ventricular myocytes. Am J Physiol 262, C75–C83.
Gilly WF & Armstrong CM (1982). Divalent cations and the activation kinetics of potassium channels in squid giant axons. J Gen Physiol 79, 965–996.
Herzberg IM, Trudeau MC & Robertson GA (1998). Transfer of rapid inactivation and sensitivity to the class III antiarrhythmic drug E-4031 from HERG to M-eag channels. J Physiol 511, 3–14.
, http://www.100md.com
Ho W-K, Earm YE, Lee SH, Brown HF & Noble D (1996). Voltage- and time-dependent block of delayed rectifier K+ current in rabbit sino-atrial node cells by external Ca2+ and Mg2+. J Physiol 494, 727–742.
Ho W-K, Kim I, Lee CO & Earm YE (1998). Voltage-dependent blockade of HERG channels expressed in Xenopus oocytes by external Ca2+ and Mg2+. J Physiol 507, 631–638.
Ho W-K, Kim I, Lee CO, Youm JB, Lee SH & Earm YE (1999). Blockade of HERG channels expressed in Xenopus laevis oocytes by external divalent cations. Biophys J 76, 1959–1971.
, 百拇医药
Jiang Y, Lee A, Chen J, Ruta V, Cadene M, Chait BT & MacKinnon R (2003a). X-ray structure of a voltage-dependent K+ channel. Nature 423, 33–41.
Jiang Y, Ruta V, Chen J, Lee A & MacKinnon R (2003b). The principle of gating charge movement in a voltage-dependent K+ channel. Nature 423, 42–48.
Johnson JP Jr, Balser JR & Bennett PB (1999). Enhancement of HERG K+ currents by Cd2+ destabilization of the inactivated state. Biophys J 77, 2534–2541.
, http://www.100md.com
Johnson JP Jr, Balser JR & Bennett PB (2001). A novel extracellular calcium sensing mechanism in voltage-gated potassium ion channels. J Neurosci 21, 4143–4153.
Keating MT & Sanguinetti MC (2001). Molecular and cellular mechanisms of cardiac arrhythmias. Cell 104, 569–580.
Liu J, Zhang M, Jiang M & Tseng GN (2002). Structural and functional role of the extracellular S5-P linker in the HERG potassium channel. J Gen Physiol 120, 723–737.
, 百拇医药
Liu J, Zhang M, Jiang M & Tseng GN (2003). Negative charges in the transmembrane domains of the HERG K channel are involved in the activation- and deactivation-gating processes. J Gen Physiol 121, 599–614.
McLaughlin S (1989). The electrostatic properties of membranes. Annu Rev Biophys Biophys Chem 18, 113–136.
Mitcheson JS, Chen J, Lin M, Culberson C & Sanguinetti MC (2000). A structural basis for drug-induced long QT syndrome. Proc Natl Acad Sci U S A 97, 12329–12333.
, 百拇医药
Papazian DM, Shao XM, Sech S-A, Mock AF, Huang Y & Wainstock DH (1995). Electrostatic interactions of S4 voltage sensor in Shaker K+ channel. Neuron 14, 1293–1301.
Paquette T, Clay JR, Ogbaghebriel A & Shrier A (1998). Effects of divalent cations on the E-4031-sensitive repolarization current, IKr, in rabbit ventricular myocytes. Biophys J 74, 1278–1285.
Piper DR, Hinz WA, Tallurri CK, Sanguinetti MC & Tristani-Firouzi M (2005). Regional specificity of human ether-à-go-go-related gene channel activation and inactivation gating. J Biol Chem 280, 7206–7217.
, 百拇医药
Piper DR, Varghese A, Sanguinetti MC & Tristani-Firouzi M (2003). Gating currents associated with intramembrane charge displacement in HERG potassium channels. Proc Natl Acad Sci U S A 100, 10534–10539.
Rulisek L & Vondrasek J (1998). Coordination geometries of selected transition metal ions (Co2+, Ni2+, Cu2+, Zn2+, Cd2+, and Hg2+) in metalloproteins. J Inorg Biochem 71, 115–127.
Sanchez-Chapula JA & Sanguinetti MC (2000). Altered gating of HERG potassium channels by cobalt and lanthanum. Pflugers Arch 440, 264–274.
, 百拇医药
Sanguinetti MC, Jiang C, Curran ME & Keating MT (1995). A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell 81, 299–307.
Sanguinetti MC & Jurkiewicz NK (1991). IK is comprised of two components in guinea pig atrial cells. Am J Physiol 260, H393–H399.
Sanguinetti MC & Xu QP (1999). Mutations of the S4–S5 linker alter activation properties of HERG potassium channels expressed in Xenopus oocytes. J Physiol 514, 667–675.
, 百拇医药
Seoh S-A, Sigg D, Papazian DM & Bezanilla F (1996). Voltage-sensing residues in the S2 and S4 segments of the Shaker K+ channel. Neuron 16, 1159–1167.
Silverman WR, Bannister JP & Papazian DM (2004). Binding site in Eag voltage sensor accommodates a variety of ions and is accessible in closed channel. Biophys J 87, 3110–3121.
Silverman WR, Roux B & Papazian DM (2003). Structural basis of two-stage voltage-dependent activation in K+ channels. Proc Natl Acad Sci U S A 100, 2935–2940.
, 百拇医药
Silverman WR, Tang CY, Mock AF, Huh KB & Papazian DM (2000). Mg2+ modulates voltage-dependent activation in ether-a-go-go potassium channels by binding between transmembrane segments S2 and S3. J Gen Physiol 116, 663–678.
Smith PL, Baukrowitz T & Yellen G (1996). The inward rectification mechanism of the HERG cardiac potassium channel. Nature 379, 833–836.
Song DK, Earm YE & Ho WK (1999). Blockade of the delayed rectifier K+ currents, IKr, in rabbit sinoatrial node cells by external divalent cations. Pflugers Arch 438, 147–153.
, 百拇医药
Spector PS, Curran ME, Zou A, Keating MT & Sanguinetti MC (1996). Fast inactivation causes rectification of the IKr channel. J Gen Physiol 107, 611–619.
Stuhmer W (1992). Electrophysiological recording from Xenopus oocytes. Methods Enzymol 207, 319–339.
Tang CY, Bezanilla F & Papazian DM (2000). Extracellular Mg2+ modulates slow gating transitions and the opening of Drosophila ether-a-Go-Go potassium channels. J Gen Physiol 115, 319–338.
, http://www.100md.com
Tiwari-Woodruff SK, Lin MA, Schulteis CT & Papazian DM (2000). Voltage–dependent structural interactions in the shaker K+ channel. J Gen Physiol 115, 123–138.
Tiwari-Woodruff SK, Schulteis CT, Mock AF & Papazian DM (1997). Electrostatic interactions between transmembrane segments mediate folding of Shaker K+ channel subunits. Biophys J 72, 1489–1500.
Torres AM, Bansal PS, Sunde M, Clarke CE, Bursill JA, Smith DJ, Bauskin A, Breit SN, Campbell TJ, Alewood PF, Kuchel PW & Vandenberg JI (2003). Structure of the HERG K+ channel S5P extracellular linker: role of an amphipathic alpha-helix in C-type inactivation. J Biol Chem 278, 42136–42148.
, http://www.100md.com
Trudeau M, Warmke JW, Ganetzky B & Robertson GA (1995). HERG, a human inward rectifier in the voltage-gated potassium channel family. Science 269, 92–95.
Warmke JW & Ganetzky B (1994). A family of potassium channel genes related to eag in Drosophila and mammals. Proc Natl Acad Sci U S A 91, 3438–3442.
Zou A, Xu QP & Sanguinetti MC (1998). A mutation in the pore region of HERG K+ channels reduces rectification by shifting the voltage dependence of inactivation. J Physiol 509, 129–138., 百拇医药(David Fernandez, Azad Gha)