Mutation of colocalized residues of the pore helix and transmembrane segments S5 and S6 disrupt deactivation and modify inactivation of KCNQ
1 Department of Physiology I, University of Tübingen, D-72076 Tübingen, Germany
2 Department of Physiology and Nora Eccles Harrison Cardiovascular Research and Training Institute, University of Utah, Salt Lake City, UT 84112, USA
Abstract
KCNQ1 (Kv 7.1) -subunits and KCNE1 subunits co-assemble to form channels that conduct the slow delayed rectifier K+ current (IKs) in the heart. Mutations in either subunit cause long QT syndrome (LQTS), an inherited disorder of cardiac repolarization. Here, the functional consequences of the LQTS-associated missense mutation V310I and several nearby residues were determined. Val310 is located at the base of the pore helix of KCNQ1, two residues below the TIGYG signature sequence that defines the K+ selectivity filter. Channels were heterologously expressed in Xenopus laevis oocytes and currents were recorded using the two-microelectrode voltage-clamp technique. V310I KCNQ1 reduced IKs amplitude when co-expressed with wild-type KCNQ1 and KCNE1 subunits. Val310 was also mutated to Gly, Ala or Leu to explore the importance of amino acid side chain volume at this position. Like V310I, V310L KCNQ1 channels gated normally. Unexpectedly, V310G and V310A KCNQ1 channels inactivated strongly and did not close normally in response to membrane hyperpolarization. Based on a homology model of the KCNQ1 channel pore, we speculate that the side group of residue 310 can interact with specific residues in the S5 and S6 domains to alter channel gating. When volume of the side chain is small, the stability of the closed state is disrupted and the extent of channel inactivation is enhanced. We mutated putative interacting residues in S5 and S6 and found that mutant Leu273 and Phe340 channels also can disrupt close states and modify inactivation. Together these findings indicate the importance of a putative pore helix–S5–S6 interaction for normal KCNQ1 channel deactivation and confirm its role in KCNQ1 inactivation. Disturbance of these interactions might underly LQTS associated with KCNQ1 mutant channels.
, 百拇医药
Introduction
Several outward currents modulate repolarization of cardiac action potentials, including the slow delayed rectifier K+ current, IKs. The IKs channel is formed by co-assembly of KCNQ1 -subunits and KCNE1 subunits (Barhanin et al. 1996; Sanguinetti et al. 1996). The most prominent effects of co-expression with KCNE1 subunits is a much slower rate of activation and an increase in current magnitude compared to homotetrameric KCNQ1 channels. In addition, IKs channels have a modified single channel conductance and no detectable inactivation when compared to KCNQ1 channels (Barhanin et al. 1996; Sanguinetti et al. 1996; Romey et al. 1997; Pusch, 1998; Sesti & Goldstein, 1998; Yang & Sigworth, 1998). The molecular mechanisms of the KCNE1-mediated changes in KCNQ1 channel gating are not well understood, but it has been suggested that the subunits may interact with the pore domain of KCNQ1 subunits (Romey et al. 1997; Tai & Goldstein, 1998; Melman et al. 2004).
, 百拇医药
The structure of KCNQ1 is assumed to be similar to other voltage-gated K+ channel -subunits, having six transmembrane domains (S1–S6), a voltage sensor (S4) and a pore helix selectivity filter segment (P-loop) that connects S5 and S6. The selectivity filter is defined by a highly conserved amino acid sequence (TIGYG) that determines ion selectivity and facilitates high rates of ion conductance by mechanisms that were elegantly defined by the crystal structure of the bacterial KcsA channel (Doyle et al. 1998; Morais-Cabral et al. 2001; Zhou et al. 2001). The carbonyl oxygen atoms of these residues from four identical subunits line a narrow pore and provide four potential binding sites for dehydrated K+ ions, two of which are normally occupied at any one time (Morais-Cabral et al. 2001). The pore helices of each subunit point towards the centre of the central cavity of the channel, and through a helix dipole effect, stabilize a single hydrated K+ ion (Doyle et al. 1998). Mutations of residues located in the P-loop of voltage-gated K+ channels (e.g. hERG) have been shown to affect C-type inactivation (Smith et al. 1996). KCNQ1 channels also exhibit inactivation, and although the molecular mechanisms are not clear, it has been suggested that the pore helix is involved (Seebohm et al. 2001).
, 百拇医药
Mutations in KCNQ1 or KCNE1 subunits cause long QT syndrome (LQTS), an inherited disorder of ventricular repolarization that predisposes affected individuals to cardiac arrhythmias and sudden death. Given the functional importance of the pore helix and selectivity filter it is not surprising that many missense mutations of amino acids located in these regions of KCNQ1 (e.g. residues 300, 305–315, 317, 318 and 320) have been associated with LQTS (see Gene Connection for the Heart website: http://pc4.fsm.it:81/cardmoc). In this study, we characterized the physiological consequences of the LQTS-associated mutation V310I, located at the base of the pore helix of the KCNQ1 subunit. Val310 is located near the K+ channel signature sequence (underlined) that forms the selectivity filter: VTTIGYG.
, 百拇医药
As expected, we previously found that channels formed by co-assembly of V310I KCNQ1, wild-type (WT) KCNQ1 and KCNE1 subunits reduced IKs amplitude (Westenskow et al. 2004). However, this mutation had no obvious effects on channel gating. To explore further the importance of Val310, we mutated this residue to other hydrophobic amino acids. The other known KCNQ channels (KCNQ2–5) have a Leu in the position homologous to Val310 of KCNQ1. Here we show that substitution of Val310 in KCNQ1 with a larger residue (Leu) did not significantly alter gating, whereas substitution with smaller amino acids (Gly or Ala) dramatically increased inactivation and prevented normal channel deactivation in response to membrane hyperpolarization. Structural modelling suggests that the volume of the side chain of residue 310 in the KCNQ1 channel might alter interactions with nearby residues located in the S5 and S6 domains to affect gating associated with both inactivation and deactivation. Further mutation of residues which putatively interact with residue Val310 was performed. The mutation of Phe273 in S5 and Phe340 in S6 can disrupt channel closure and modify inactivation. Together these findings indicate the importance of the pore helix for normal channel deactivation and confirm its role in C-type inactivation. Impairment of the interactions might underly LQTS associated with KCNQ1 mutant channels.
, 百拇医药
Methods
Molecular biology
Site-directed mutagenesis of human KCNQ1 (KVLQT1) was performed by PCR using the megaprimer method (Sarkar & Sommer, 1990) and native Pfu-polymerase. All constructs were confirmed by automated DNA sequencing. The constructs were linearized with NheI and cRNA was transcribed in vitro with Capscribe and T7 polymerase (Roche Molecular Biochemicals). Human KCNE1 (minK) was linearized with EcoRI and cRNA was made using Capscribe and SP6 polymerase.
, 百拇医药
Oocyte isolation and two-electrode voltage clamp
Ovarian lobes were harvested from Xenopus laevis frogs anaesthetized with a 0.17% tricaine solution. After surgery frogs were allowed to recover and after the final oocyte collection, frogs were killed humanely. The use of frogs was approved by the University of Utah Institutional Animal Care and Use Committee. Follicle cells were removed from the oocytes by treatment for 90–150 min with collagenase (1 mg ml–1, Worthington, type II) in ND96-Ca2+-free solution containing (mM): NaCl 96, KCl 2, MgCl2 1 and Hepes 5 (pH 7.6). Oocytes were subsequently stored at 18°C in Barth's solution containing (mM): NaCl 88, KCl 1, CaCl2 0.4, MgCl2 1 and Hepes 10 (pH 7.4) plus gentamycin (50 mg l–1). For characterization of homomeric KCNQ1 channels, each oocyte was injected with 10–15 ng of WT or mutant KCNQ1 cRNA. Oocytes were co-injected with cRNA encoding WT KCNQ1 (6 ng) and KCNE1 (0.15 ng), or WT KCNQ1 (3 ng) plus V310I KCNQ1 (3 ng) plus KCNE1 (0.15 ng) to determine functional effects of mutant subunits on IKs. The basic recording solution contained (mM): NaOH 96, KOH 2, Ca(OH)2 2, MgCl2 1, Mes 101, and Hepes 5; and was equilibrated to pH 7.6. This solution was altered for the ion selectivity experiments by replacing NaOH by KOH as appropriate to maintain a constant concentration of monovalent cations.
, 百拇医药
A Dagan TEV-200 amplifier or a NPI TurboTEC-10CX amplifier was used to record currents at 23°C in oocytes 2–3 days after injection with cRNA using standard two-electrode voltage-clamp techniques. Data acquisition was performed using a Pentium IV computer, a Digidata 1322 A/D interface and pClamp ver. 8 software (Axon Instruments). Specific pulse protocols used to voltage clamp oocytes are described in Results.
Data analysis
Analysis of data was performed with Clampfit 8 (Axon Instruments) and Origin 6 or 7 (Microcal) software. Briefly, the decaying phase of tail currents were fitted with a single exponential function and the fit was extrapolated to the beginning of the repolarization pulse. Extrapolated tail current amplitudes were normalized to the values determined after a pulse to +40 mV and the resulting data were fitted to a Boltzmann function:
, 百拇医药
where I is current amplitude, Imax is maximal tail current, Vt is the test voltage, V0.5 is the half-point of activation and k is the slope factor of the condcutance (G)–V relationship. G was calculated by measuring the current amplitude at the end of a 2-s pulse and dividing this value by the test potential minus the reversal potential (Vt–Erev) determined for each oocyte.
At –120 mV, WT KCNQ1 channels fully deactivated to a non-conducting closed state, whereas some of the mutant channels failed to close completely. To estimate the extent of channel closure at –120 mV, the tail currents elicited at this voltage after a step to 40 mV were fitted to a bi-exponential function. The slow component, representing deactivation, was extrapolated back to the beginning of the tail pulse and its amplitude at this point gave a value for the closing component a. In some cases mono-exponential fits were sufficient to describe deactivation because no hook in the tail was obvious. The difference between the steady-state non-deactivating component and zero current gave the value b. The calculated b/a ratios were used to estimate the relative channel open probability (Popen) at –120 mV.
, 百拇医药
Activating and deactivating current traces were fitted using a simplex algorithm to one (eqn (2)) or two (eqn (3)) exponential functions:
where y is current, A is current amplitude, t is time and is the time constant for the rise or decay of current.
Student's t test or ANOVA was used to test for statistical significance between groups. A value of P < 0.05 was considered statistically significant. Numerical values are reported as means ± S.E.M. (n = number of experiments).
, http://www.100md.com
Molecular modelling of the KCNQ1 channel pore
A 3-dimensional model of the S5 to S6 domains of the KCNQ1 subunit was constructed based on homology with the crystal structure of the corresponding domains of KcsA, a proton-activated K+ channel isolated from Streptomyces lividans (Doyle et al. 1998). The sequence of KCNQ1 is 46% homologous and 36% identical to KcsA within these regions. Initial models of KCNQ1 were built using Swiss-Model (Glaxo Welcome – http://www.expasy.org/swissmod/SWISS-MODEL.html) and energy minimized using GROMOS 96 (http://igc.ethz.ch/gromos) in the default configuration implemented in Swiss-Model.
, 百拇医药
Results
Kinetic analysis of WT and V310I KCNQ1 currents
Channels formed by homomultimeric assembly of subunits were characterized in oocytes injected with either WT KCNQ1 or V310I KCNQ1 cRNA. WT and mutant channels were expressed in oocytes and currents elicited by 2-s voltage-clamp pulses to potentials ranging from –100 to +60 mV from a holding potential of –70 mV. Both WT and V310I KCNQ1 channel currents activated with a multi-exponential time course in response to membrane depolarization and did not fully activate during a 2-s pulse (Fig. 1A and B). At potentials of +40 mV and higher the fast component of activation was followed by a transient reduction (due to faster rate of inactivation) and a subsequent second slower phase of activation. Tail currents were measured at –60 mV and were characterized by an initial increase in amplitude as channels recovered from inactivation, followed by a slower decay as channels deactivated.
, 百拇医药
A, WT KCNQ1 channel current elicited by 2-s pulses to potentials ranging from –100 to +60 mV, applied in 10 mV steps from a holding potential of –70 mV. Inset illustrates the voltage pulse protocol. B, V310I KCNQ1 channel currents elicited with the same protocol described for WT channels. For A and B, the calibration bars represent 1 s and 1 μA, respectively. C, relative conductance–voltage relationship for WT KCNQ1 channel currents () and V310I KCNQ1 currents () (n = 6–8). D, time constant of the fast activating component () plotted against test voltage for WT () and V310I () KCNQ1 currents (n = 5–8). E, time constants of deactivation () plotted against test voltage for WT () and V310I () KCNQ1 currents (n = 5–8). F, time constants for recovery from inactivation as a function of test potential for WT () and V310I () KCNQ1 currents (n = 5–8).
, 百拇医药
The voltage dependence of activation was determined from an analysis of tail currents elicited after 2-s activating pulses. The V0.5 was –21.7 mV and the slope factor was 13.9 mV for the voltage dependence of activation of WT KCNQ1 channels (n = 6). The V0.5 was –21.5 mV and the slope factor was 13.7 mV for V310I KCNQ1 channels (n = 6). The voltage dependence of WT and mutant KCNQ1 channel conductance was also estimated by measuring the amplitude of currents at the end of 2-s pulses applied to a variable potential, normalized to the driving force for K+ and plotted versus the test voltage (Fig. 1C). Peak V310I KCNQ1 conductance saturated near +60 mV, whereas the conductance of WT KCNQ1 progressively decreased at voltages above +20 mV, presumably due to increased channel inactivation. Although the conductance of V310I was altered, the kinetics of fast activation (Fig. 1D) was similar to WT channels. Deactivation was preceded by a transient increase in current, caused by rapid recovery of channels from inactivation in both WT and V310I channel currents. Tail currents elicited after a 2-s pulse to +40 mV were fitted to a bi-exponential function, where the fast and slow components reflected recovery from inactivation and subsequent deactivation, respectively. Neither measure of gating (recovery from inactivation or deactivation) was significantly altered by the V310I mutation (Fig. 1E and F).
, http://www.100md.com
Co-expression of WT and mutant KCNQ1 subunits with KCNE1
We recently reported that V310I currents are about 60% smaller than WT KCNQ1 currents and that IKs induced by co-expressing WT KCNE1 plus V310I KCNQ1 subunits was 89% smaller that WT IKs (Westenskow et al. 2004). Here, we extend the evaluation of V310I to include functional effects of co-expressing the different subunits in a ratio chosen to mimic the presumed expression pattern of an individual harbouring a heterozygous mutation. Oocytes were co-injected with cRNA encoding WT KCNQ1 (6 ng) and KCNE1 (0.15 ng), or WT KCNQ1 (3 ng) plus V310I KCNQ1 (3 ng) plus KCNE1 (0.15 ng) to determine functional effects of mutant subunits on IKs. KCNE1 cRNA (0.15 ng) was also injected alone to define the ‘background’ IKs which results from the interaction of exogenous KCNE1 and constitutively expressed KCNQ1 subunits (Sanguinetti et al. 1996). Examples of currents induced by injection of cRNAs are illustrated in Fig. 2A–C, and the current–voltage relationships for these experiments are summarized in Fig. 2D. The currents recorded in oocytes co-expressing KCNE1, WT KCNQ1 and V310I KCNQ1 were kinetically indistinguishable from oocytes expressing KCNE1 and WT KCNQ1 subunits alone. However, the magnitude of current was reduced by the presence of the mutant KCNQ1 subunit.
, 百拇医药
A–C, currents recorded from Xenopus oocytes injected with 6 ng WT KCNQ1 cRNA plus 0.15 ng KCNE1 cRNA (A), 0.15 ng KCNE1 cRNA alone (B) or 3 ng WT KCNQ1 cRNA plus 3 ng V310I KCNQ1 cRNA plus 0.15 ng KCNE1 cRNA (C). The scale bars in A represent 1 s and 1 μA. D, current–voltage relationships for currents measured at the end of 7-s pulses to the indicated test potentials.
Effects of Val310 mutations depend on volume of the substituted amino acid side group
, 百拇医药
The only obvious effects of the V310I mutation were to reduce the extent of inactivation of KCNQ1 and to reduce the amplitude of IKs. To explore further the potential influence of Val310 on the gating properties of KCNQ1 channels, this residue was mutated to Gly, Ala and Leu. These residues were chosen because of their similarity to Val and Ile with respect to hydrophobicity and molecular volume compared to the other natural amino acids. The mutant channels were expressed in oocytes and currents elicited by the same voltage pulse protocol described for WT channels in Fig. 1A. Representative traces of WT and mutant channel currents are shown in Fig. 3. The rate of current activation for the mutant channels was similar to WT channels. However, mutation of Val310 to smaller residues (Ala or Gly) greatly enhanced the rate of channel inactivation and prevented channels from fully closing (Fig. 3B and C). Tail currents were too complex to analyse kinetics, but they appeared to be hooked, as expected if channels recovered from inactivation before deactivation when the cell was repolarized to –60 mV. Channels containing a mutation of Val310 to a larger residue (Leu) activated and inactivated in a similar manner to WT or V310I KCNQ1, but the tail currents did not exhibit a hook. This does not necessarily imply that V310L channels have reduced inactivation compared to WT channels. Indeed, currents induced during depolarizing pulses to +30 mV exhibited a clear transient phase, suggesting channel inactivation. Instead, the lack of hooked tail currents suggests that recovery from inactivation was either too fast to observe, or very slow compared to the rate of deactivation. The normalized peak current–voltage relationship for V310L KCNQ1 was very similar to WT KCNQ1 channels (Fig. 4A). The V0.5 for the voltage dependence of activation for V310L KCNQ1 was –22.7 mV with a slope factor of 13.5 mV (n = 7). Because of extensive inactivation, the conductance–voltage relationships for V310A and V310G KCNQ1 (Fig. 4B) were determined from peak current values rather than from the end of the 2-s pulse. The current–voltage relationships were determined for both peak currents and currents at the end of 2-s test pulses (Fig. 4C and D). These current–voltage plots clearly show the extent of inactivation induced by the V310A and V310G mutations and that the channels did not close at potentials as negative as –100 mV. The ratio of current at the end of the pulse divided by the peak current for the test pulse to +40 mV was 0.14 for V310A and 0.26 for V310G KCNQ1. Thus, mutation of Val310 to smaller residues accentuated inactivation and appeared to prevent channel deactivation. However, as discussed below, V310G and V310A channels partially deactivated when tail currents were elicited at potentials more negative than –100 mV.
, http://www.100md.com
A, WT KCNQ1 currents and voltage pulse protocol. B–E, currents recorded from oocytes expressing the indicated Val310 mutant channels. In each panel, the scale bars represent 1 s and 1 μA.
A, conductance–voltage relationship for WT, V310I and V310L KCNQ1 (n = 6–7). B, peak currents plotted against the test pulse potential for V310G and V310A KCNQ1 channel currents (n = 6–8). C and D, peak currents for V310G and V310A KCNQ1 channels and currents measured at the end of 2-s test pulses to the indicated potentials (n = 6–8).
, 百拇医药
Val310 mutant channels deactivate faster than WT channels but fail to close normally
Tail currents for WT and mutant KCNQ1 channels were elicited by first pulsing to +40 mV to activate channels then returning the membrane to potentials ranging from –120 to –20 mV, applied in 20 mV increments. The rate of deactivation was slow for WT (Fig. 5A) and enhanced by mutations of Val310 (Fig. 5B–E). The fraction of channels that remained open at –120 mV (relative open probability, Popen) was calculated as described in Methods. Relative Popen was 0.02 ± 0.001 for WT KCNQ1 (n = 7) and 0.03 ± 0.001 (n = 7) for V310I KCNQ1. However, a small fraction of V310L channels remained open at –120 mV (Popen = 0.10 ± 0.01, n = 7), as did a substantial fraction of V310G channels (Popen = 0.32 ± 0.001, n = 6) and V310A channels (Popen = 0.48 ± 0.01, n = 13; Fig. 5F).
, 百拇医药
A–E, WT and mutant channel currents elicited by 2-s pulses, 125-ms pulses (V310G) or 300-ms pulses to +40 mV followed by pulses to different potentials as indicated by inset between A and B. F, relative open probability (Popen) determined at –120 mV for WT and mutant channels (n = 6–13).
Selective permeability of mutant channels
The relative cation selectivity of K+ channels is determined by the selectivity filter. Thus, based solely on the location of Val310, it is conceivable that mutations of this residue also could alter cation selectivity. To test for this possibility, the reversal potential (Erev) for KCNQ1 currents was determined for oocytes bathed in solutions containing an extracellular [K+] of 4, 16, 32 or 100 mM. Linear fits of the averaged data had slopes of 51–56 mV per 10-fold increase in [K+]o (Fig. 6), indicating that mutant channels maintained relatively normal K+ selectivity.
, http://www.100md.com
The reversal potential (Erev) for tail currents illustrated in Fig. 5 were measured and plotted against the extracellular [K+] for WT and mutant KCNQ1 channels expressed in oocytes. The slopes of the linear fits were 51–56 mV per 10-fold increase of extracellular [K+] (n = 6–8).
Residue Val310 putatively interacts with transmembrane helices S5 and S6
A 3-dimensional model of the pore domain of KCNQ1 was generated based on the high homology to the closed channel structure of the KcsA potassium channel (Fig. 7). The amino acid Val310 probably interacts with transmembrane helices S5 and S6, more specifically with Leu273 (S5), Phe332 (S6), Ala336 (S6) and Phe340 (S6). The residues were mutated and resulted in exchanges of the residue: Leu273 to Ala, Val, Phe or Gln, residue Phe332 to Ala, residue Ala336 to Gly or Cys and residue Phe340 to Ala, Ile or Tyr. These channels were expressed and L273A and L273Q were found to be non-functional. Further, F332A was very weakly functional. These channels could therefore not be studied and analysed in detail. Whereas all Phe340 mutant channels inactivated strongly the two functional Leu273 channels had a very variable impact on inactivation of channels (almost no inactivation in L273V and very strong inactivation in L273F). The tail currents of all other mutant channels were analysed and the relative open probability Popen was calculated. The Ala336 mutants closed normally. The mutant Leu273 and Phe344 channels could not close completely. However, the side chain volume at Leu273 did not matter (n = 6–8). The side chain volume of residue Phe340 determined Popen to some extend (Fig. 8). The larger residue Tyr at 340 allowed 83% (n = 8) of the channels to close whereas the smaller residues Ile and Ala allowed only 51% (n = 8) and 70% (n = 9) of channels, respectively, to close completely.
, 百拇医药
A KCNQ1 homology model was generated for the S5–S6 region based on the known KcsA crystal structure. The model shows residues Thr265–Tyr281 in the S5 domain in green and residues Trp323–Ile346 in the S6 domain in orange, all in a space-fill representation. The model was rotated 45 deg in the direction indicated by the arrows. The pore helix and the linkers are depicted as blue ribbons. Val310 is coloured red and residues which are in close proximity are labelled and coloured lighter than residues not in proximity to Val310.
, 百拇医药
A, representative current traces of mutant KCNQ1 channels. The current–voltage tracings were elicited with pulse protocols as used in Fig. 1. Mutant channel currents were further elicited by a 3-s pulse to +40 mV followed by pulses to potentials of –140 mV to –40 mV in 20 mV increments. The scale bars at the current–voltage traces represent 1 μA and 1 s. The current traces showing channel deactivation are scaled to about the same size to allow better view on the tail currents. B, relative open probability (Popen) determined at –120 mV for mutant channels (n = 5–9). n.f., non-functional. xb
, 百拇医药
Discussion
Most LQTS-associated missense mutations in KCNQ1 cause loss of function when the mutant subunits are expressed alone, and a dominant-negative effect when co-expressed with WT KCNQ1 and KCNE1 subunits (Shalaby et al. 1997). V310I is another example of a mutation that causes partial loss of function of IKs. Similar to previous studies (Barhanin et al. 1996; Sanguinetti et al. 1996), we found that current induced by injection of oocytes with KCNQ1 and KCNE1 cRNA was 25-fold larger than that induced by KCNE1 alone. In contrast, the average current induced by co-expression of V310I KCNQ1, WT KCNQ1 and KCNE1 subunits was 15-fold larger than that induced by KCNE1 alone. These data indicate that co-assembly of KCNQ1 and KCNE1 with V310I KCNQ1 subunits suppressed channel function, an effect that could result from a reduced single channel conductance, a higher flicker/block rate (Pusch et al. 2001), dysfunction of protein trafficking or protein misfolding and an accelerated rate of degradation. Whatever the exact mechanism, a decrease in repolarizing current would prolong action potential duration and QT interval, and increase the risk of ventricular arrhythmias.
, 百拇医药
The functional consequence of the V310I mutation was a simple loss of function. However, substitution of Val310 with smaller residues had a profound effect on channel gating. Mutation of Val310 to Gly or Ala greatly enhanced inactivation and prevented normal channel closure, whereas mutation to a larger residue (Leu or Ile) had only minor effects on channel gating. The relative hydrophobicity values (normalized to a value of zero for Gly) for the side chains of Ala, Val, Leu and Ile are –0.87, –3.10, –3.98 and –3.98. The accessible surface areas (ASA) for the same residues are 85, 113, 160, 180 and 182 2, respectively (Crighton, 1993). The median ASA for the residues in -helical proteins are 3, 5, 3, 4 and 4 suggesting that these hydrophobic residues are mostly hidden in the proteins (Lins et al. 2003). The gating of the V310G and V310A mutant channels were affected similarly, suggesting that size of the side chain was a factor contributing to the observed gating defects. KCNQ2–5 channels have a Leu in a position homologous to Val310 of KCNQ1. Mutation of Val310 to Leu in KCNQ1 eliminated the hooks normally observed in the tail currents, an effect that might be due to a faster rate of recovery from inactivation compared to channel deactivation.
, 百拇医药
Inactivation of voltage-gated K+ channels can occur by N-type (Hoshi et al. 1991), C-type (Lopez-Barneo et al. 1993) or P-type (De Biasi et al. 1993) mechanisms, named in reference to the structural region of the channel believed to participate in the gating mechanism. P refers to the pore domain and is often lumped together with C-type inactivation. It has been suggested that P-type inactivation involves a slight constriction of the narrow canal formed by the selectivity filter, causing a transient increase in the permeability of Na+ relative to K+ before the channel ceases to conduct (Kiss & Korn, 1998). C-type inactivation represents a stabilized P-type inactivated state of the channel (Loots & Isacoff, 1998). Perhaps reducing the volume of the side chain of residue 310 enhances the ability of the selectivity filter to constrict, leading to enhanced voltage-dependent inactivation. The location of residue 310 at the end of the pore helix closest to the selectivity filter limits the number of possible interactions with neighbouring residues. A homology model of the S5–S6 domains of KCNQ1 (Fig. 7), based on the structure of KcsA, reveals several residues that are in close proximity to Val310. Besides the immediately adjacent residues (Thr309 and Thr311), Val310 is in close contact with Leu273 of S5 and Ala336, Ile337 and Phe340 of S6. We mutated these residues and found that mutation of Leu273 to Val or Phe and Phe340 to small residues disrupted channel closure. Small residues would disrupt interaction of Phe340 side chain with Val310 based on modelling and together with the experimental data interaction of Phe340 with Val310 seems likely.
, 百拇医药
The LQTS-associated mutation L273F caused enhanced inactivation of KCNQ1 homomeric channels (Shalaby et al. 1997; Seebohm et al. 2001), an effect similar to that observed here for mutation of Val310 to Ala or Gly. Mutation of Leu273 to Val produced a weakly inactivating channel. Thus, the Val310 interaction with residue Leu273 does not only allow correct channel closure; it is also of high importance for channel inactivation. Perhaps a hydrophobic interaction between L273 of the S5 domain and Val310 of the pore helix stabilizes the open state of the selectivity filter (Liu & Joho, 1998). Disruption of hydrophobic interactions in the Val310/Leu273/Phe340 cluster by mutation of Val310 to the neutral amino acids Ala or Gly (or mutation of Leu273 to Phe or Phe340 to small or hydrophilic amino acids) might destabilize the open state of the pore, causing an enhanced rate and extent of inactivation. Disruption of these interactions by V310I and L273F might cause LQTS associated with KCNQ1 mutant channels. It seems plausible that nearby mutations also influence the Val310–Leu273–Phe340 interactions. In fact, every single amino acid in the lower pore helix (305–312) and several in S5 (266, 269, 273, 275–277) and S6 (341, 342, 344, 345) result in reduced function (summary: http://pc4.fsm.it:81/cardmoc/) and LQTS.
, http://www.100md.com
Given the suspected importance of the pore helix and surrounding regions in the process of P- and C-type inactivation, perhaps it was not too surprising to discover that mutations of Val310 to other amino acids could affect inactivation. However, the finding that mutation of this residue to Gly or Ala impaired channel closure was totally unexpected. The mechanism of incomplete deactivation in these mutant channels is unknown, but may relate to the relative ability of the S6 domain to bend back and forth at the proposed hinge point for the activation gate. The hinge for the activation gate of the bacterial K+ channel MthK is a Gly residue located near the midpoint of the S6 domain (Jiang et al. 2002). This Gly is conserved in most, but not all Kv channels. In KCNQ1, this residue is an Ala (Ala336), located close to residue 310 (Fig. 7). Mutation of Val310 to a smaller residue (Gly or Ala) may not prevent the full range of bending and/or twisting of S6 required for normal closure of the activation gate. Indeed, we have studied the impact of the Gly at the putative glycine hinge (KCNQ1 Ala336) and found it to be of minor importance for KCNQ1 channel gating (authors' unpublished observations). It is more likely that the lack of channel closure might be due to disruption of a pore helix–S6 interaction similar to that previously described for Kv2.1 channels. It was proposed that interaction of a residue near the base of the pore helix (Thr370) with a S6 residue (Cys393) in Kv2.1 stabilizes the open state and that counter-clockwise rotation of S6 at negative transmembrane potentials disrupts this interaction, allowing the channel to close (Liu & Joho, 1998). Val310 of KCNQ1 is located next to the residue homologous to Thr370 of Kv2.1, and Phe332 of KCNQ1 is in the same position as C393 of Kv2.1. Mutation of a single S6 residue (Cys480) in exp-2 channels of C. elegans has also been associated with disrupted deactivation. This affect was attributed to inability of the mutant S6 domain to rotate normally during gating due to a steric hindrance of the bulky aromatic group with specific residues in the adjoining S5 domain (Liu & Joho, 1998; Espinosa et al. 2001). Because KCNQ1 is not strictly dependent on a gating hinge (authors' unpublished observations) the proposed gating of KCNQ1 might include straighter S5 and S6 helices compared to the pronounced gating hinge in MthK. It seems possible that the central important positioning of the pore helix, S5 and S6 have a strong impact on channel state stability.
, http://www.100md.com
In summary, mutation of Val310, a residue located at the base of the pore helix of KCNQ1, can alter channel gating associated with inactivation and deactivation. Homology modelling and experimental data suggest a complex interaction between residue 310 and closely situated residues in the pore helix, S5 (Leu273) and S6 (Phe340). These interactions may determine the stability of two gates, one associated with channel activation and deactivation and formed by criss-crossing of the intracellular ends of the S6 domains (Doyle et al. 1998), and another one associated with P/C-type inactivation and formed by the selectivity filter–pore helix complex (Loots & Isacoff, 1998). Disruption of these interactions might underly LQTS associated with KCNQ1 mutant channels V310I and L273F and possibly channels with nearby mutations.
, 百拇医药
References
Barhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M & Romey G (1996). K(V)LQT1 and lsK (minK) proteins associate to form the I Ks cardiac potassium current. Nature 384, 78–80.
Crighton TE (1993). Proteins Structures and Molecular Properties 2nd edn. HW Freeman, New York.
De Biasi M, Drewe JA, Kirsch GE & Brown AM (1993). Histidine substitution identifies a surface position and confers Cs+ selectivity on a K+ pore. Biophys J 65, 1235–1242.
, 百拇医药
Doyle DA, Morais CJ, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT & MacKinnon R (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69–77.
Espinosa F, Fleischhauer R, McMahon A & Joho RH (2001). Dynamic interaction of S5 and S6 during voltage-controlled gating in a potassium channel. J Gen Physiol 118, 157–170.
Hoshi T, Zagotta WN & Aldrich RW (1991). Two types of inactivation in Shaker K+ channels: effects of alterations in the carboxy-terminal region. Neuron 7, 547–556.
, 百拇医药
Jiang Y, Lee A, Chen J, Cadene M, Chait BT & MacKinnon R (2002). The open pore conformation of potassium channels. Nature 417, 523–526.
Kiss L & Korn SJ (1998). Modulation of C-type inactivation by K+ at the potassium channel selectivity filter. Biophys J 74, 1840–1849.
Lins L, Thomas A & Brasseur R (2003). Analysis of accessible surface of residues in proteins. Protein Sci 12, 1406–1417.
, 百拇医药
Liu Y & Joho RH (1998). A side chain in S6 influences both open-state stability and ion permeation in a voltage-gated K+ channel. Pflugers Arch 435, 654–661.
Loots E & Isacoff EY (1998). Protein rearrangements underlying slow inactivation of the Shaker K+ channel. J Gen Physiol 112, 377–389.
Lopez-Barneo J, Hoshi T, Heinemann SH & Aldrich RW (1993). Effects of external cations and mutations in the pore region on C-type inactivation of Shaker potassium channels. Receptors Channels 1, 61–71.
, http://www.100md.com
Melman YF, Um SY, Krumerman A, Kagan A & McDonald TV (2004). KCNE1 binds to the KCNQ1 pore to regulate potassium channel activity. Neuron 42, 927–937.
Morais-Cabral JH, Zhou Y & MacKinnon R (2001). Energetic optimization of ion conduction rate by the K+ selectivity filter. Nature 414, 37–42.
Pusch M (1998). Increase of the single-channel conductance of KvLQT1 potassium channels induced by the association with minK. Pflugers Arch 437, 172–174.
, 百拇医药
Pusch M, Ferrera L & Friedrich T (2001). Two open states and rate-limiting gating steps revealed by intracellular Na+ block of human KCNQ1 and KCNQ1/KCNE1 K+ channels. J Physiol 533, 135–143.
Romey G, Attali B, Chouabe C, Abitbol I, Guillemare E, Barhanin J & Lazdunski M (1997). Molecular mechanism and functional significance of the MinK control of the KvLQT1 channel activity. J Biol Chem 272, 16713–16716.
Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS, Atkinson DL & Keating MT (1996). Coassembly of K(V)LQT1 and minK (IsK) proteins to form cardiac IKs potassium channel. Nature 384, 80–83.
, 百拇医药
Sarkar G & Sommer SS (1990). The ‘megaprimer’ method of site-directed mutagenesis. Biotechniques 8, 404–407.
Seebohm G, Scherer CR, Busch AE & Lerche C (2001). Identification of specific pore residues mediating KCNQ1 inactivation. A novel mechanism for long QT syndrome. J Biol Chem 276, 13600–13605.
Sesti F & Goldstein SA (1998). Single-channel characteristics of wild-type IKs channels and channels formed with two minK mutants that cause long QT syndrome. J Gen Physiol 112, 651–663.
, http://www.100md.com
Shalaby FY, Levesque PC, Yang WP, Little WA, Conder ML, Jenkins-West T & Blanar MA (1997). Dominant-negative KvLQT1 mutations underlie the LQT1 form of long QT syndrome. Circulation 96, 1733–1736.
Smith PL, Baukrowitz T & Yellen G (1996). The inward rectification mechanism of the HERG cardiac potassium channel. Nature 379, 833–836.
Tai KK & Goldstein SA (1998). The conduction pore of a cardiac potassium channel. Nature 391, 605–608.
, 百拇医药
Westenskow P, Splawski I, Timothy KW, Keating MT & Sanguinetti MC (2004). Compound mutations: a common cause of severe long-QT syndrome. Circulation 109, 1834–1841.
Yang Y & Sigworth FJ (1998). Single-channel properties of IKs potassium channels. J Gen Physiol 112, 665–678.
Zhou Y, Morais-Cabral JH, Kaufman A & MacKinnon R (2001). Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution. Nature 414, 43–48., http://www.100md.com(Guiscard Seebohm, Peter W)
2 Department of Physiology and Nora Eccles Harrison Cardiovascular Research and Training Institute, University of Utah, Salt Lake City, UT 84112, USA
Abstract
KCNQ1 (Kv 7.1) -subunits and KCNE1 subunits co-assemble to form channels that conduct the slow delayed rectifier K+ current (IKs) in the heart. Mutations in either subunit cause long QT syndrome (LQTS), an inherited disorder of cardiac repolarization. Here, the functional consequences of the LQTS-associated missense mutation V310I and several nearby residues were determined. Val310 is located at the base of the pore helix of KCNQ1, two residues below the TIGYG signature sequence that defines the K+ selectivity filter. Channels were heterologously expressed in Xenopus laevis oocytes and currents were recorded using the two-microelectrode voltage-clamp technique. V310I KCNQ1 reduced IKs amplitude when co-expressed with wild-type KCNQ1 and KCNE1 subunits. Val310 was also mutated to Gly, Ala or Leu to explore the importance of amino acid side chain volume at this position. Like V310I, V310L KCNQ1 channels gated normally. Unexpectedly, V310G and V310A KCNQ1 channels inactivated strongly and did not close normally in response to membrane hyperpolarization. Based on a homology model of the KCNQ1 channel pore, we speculate that the side group of residue 310 can interact with specific residues in the S5 and S6 domains to alter channel gating. When volume of the side chain is small, the stability of the closed state is disrupted and the extent of channel inactivation is enhanced. We mutated putative interacting residues in S5 and S6 and found that mutant Leu273 and Phe340 channels also can disrupt close states and modify inactivation. Together these findings indicate the importance of a putative pore helix–S5–S6 interaction for normal KCNQ1 channel deactivation and confirm its role in KCNQ1 inactivation. Disturbance of these interactions might underly LQTS associated with KCNQ1 mutant channels.
, 百拇医药
Introduction
Several outward currents modulate repolarization of cardiac action potentials, including the slow delayed rectifier K+ current, IKs. The IKs channel is formed by co-assembly of KCNQ1 -subunits and KCNE1 subunits (Barhanin et al. 1996; Sanguinetti et al. 1996). The most prominent effects of co-expression with KCNE1 subunits is a much slower rate of activation and an increase in current magnitude compared to homotetrameric KCNQ1 channels. In addition, IKs channels have a modified single channel conductance and no detectable inactivation when compared to KCNQ1 channels (Barhanin et al. 1996; Sanguinetti et al. 1996; Romey et al. 1997; Pusch, 1998; Sesti & Goldstein, 1998; Yang & Sigworth, 1998). The molecular mechanisms of the KCNE1-mediated changes in KCNQ1 channel gating are not well understood, but it has been suggested that the subunits may interact with the pore domain of KCNQ1 subunits (Romey et al. 1997; Tai & Goldstein, 1998; Melman et al. 2004).
, 百拇医药
The structure of KCNQ1 is assumed to be similar to other voltage-gated K+ channel -subunits, having six transmembrane domains (S1–S6), a voltage sensor (S4) and a pore helix selectivity filter segment (P-loop) that connects S5 and S6. The selectivity filter is defined by a highly conserved amino acid sequence (TIGYG) that determines ion selectivity and facilitates high rates of ion conductance by mechanisms that were elegantly defined by the crystal structure of the bacterial KcsA channel (Doyle et al. 1998; Morais-Cabral et al. 2001; Zhou et al. 2001). The carbonyl oxygen atoms of these residues from four identical subunits line a narrow pore and provide four potential binding sites for dehydrated K+ ions, two of which are normally occupied at any one time (Morais-Cabral et al. 2001). The pore helices of each subunit point towards the centre of the central cavity of the channel, and through a helix dipole effect, stabilize a single hydrated K+ ion (Doyle et al. 1998). Mutations of residues located in the P-loop of voltage-gated K+ channels (e.g. hERG) have been shown to affect C-type inactivation (Smith et al. 1996). KCNQ1 channels also exhibit inactivation, and although the molecular mechanisms are not clear, it has been suggested that the pore helix is involved (Seebohm et al. 2001).
, 百拇医药
Mutations in KCNQ1 or KCNE1 subunits cause long QT syndrome (LQTS), an inherited disorder of ventricular repolarization that predisposes affected individuals to cardiac arrhythmias and sudden death. Given the functional importance of the pore helix and selectivity filter it is not surprising that many missense mutations of amino acids located in these regions of KCNQ1 (e.g. residues 300, 305–315, 317, 318 and 320) have been associated with LQTS (see Gene Connection for the Heart website: http://pc4.fsm.it:81/cardmoc). In this study, we characterized the physiological consequences of the LQTS-associated mutation V310I, located at the base of the pore helix of the KCNQ1 subunit. Val310 is located near the K+ channel signature sequence (underlined) that forms the selectivity filter: VTTIGYG.
, 百拇医药
As expected, we previously found that channels formed by co-assembly of V310I KCNQ1, wild-type (WT) KCNQ1 and KCNE1 subunits reduced IKs amplitude (Westenskow et al. 2004). However, this mutation had no obvious effects on channel gating. To explore further the importance of Val310, we mutated this residue to other hydrophobic amino acids. The other known KCNQ channels (KCNQ2–5) have a Leu in the position homologous to Val310 of KCNQ1. Here we show that substitution of Val310 in KCNQ1 with a larger residue (Leu) did not significantly alter gating, whereas substitution with smaller amino acids (Gly or Ala) dramatically increased inactivation and prevented normal channel deactivation in response to membrane hyperpolarization. Structural modelling suggests that the volume of the side chain of residue 310 in the KCNQ1 channel might alter interactions with nearby residues located in the S5 and S6 domains to affect gating associated with both inactivation and deactivation. Further mutation of residues which putatively interact with residue Val310 was performed. The mutation of Phe273 in S5 and Phe340 in S6 can disrupt channel closure and modify inactivation. Together these findings indicate the importance of the pore helix for normal channel deactivation and confirm its role in C-type inactivation. Impairment of the interactions might underly LQTS associated with KCNQ1 mutant channels.
, 百拇医药
Methods
Molecular biology
Site-directed mutagenesis of human KCNQ1 (KVLQT1) was performed by PCR using the megaprimer method (Sarkar & Sommer, 1990) and native Pfu-polymerase. All constructs were confirmed by automated DNA sequencing. The constructs were linearized with NheI and cRNA was transcribed in vitro with Capscribe and T7 polymerase (Roche Molecular Biochemicals). Human KCNE1 (minK) was linearized with EcoRI and cRNA was made using Capscribe and SP6 polymerase.
, 百拇医药
Oocyte isolation and two-electrode voltage clamp
Ovarian lobes were harvested from Xenopus laevis frogs anaesthetized with a 0.17% tricaine solution. After surgery frogs were allowed to recover and after the final oocyte collection, frogs were killed humanely. The use of frogs was approved by the University of Utah Institutional Animal Care and Use Committee. Follicle cells were removed from the oocytes by treatment for 90–150 min with collagenase (1 mg ml–1, Worthington, type II) in ND96-Ca2+-free solution containing (mM): NaCl 96, KCl 2, MgCl2 1 and Hepes 5 (pH 7.6). Oocytes were subsequently stored at 18°C in Barth's solution containing (mM): NaCl 88, KCl 1, CaCl2 0.4, MgCl2 1 and Hepes 10 (pH 7.4) plus gentamycin (50 mg l–1). For characterization of homomeric KCNQ1 channels, each oocyte was injected with 10–15 ng of WT or mutant KCNQ1 cRNA. Oocytes were co-injected with cRNA encoding WT KCNQ1 (6 ng) and KCNE1 (0.15 ng), or WT KCNQ1 (3 ng) plus V310I KCNQ1 (3 ng) plus KCNE1 (0.15 ng) to determine functional effects of mutant subunits on IKs. The basic recording solution contained (mM): NaOH 96, KOH 2, Ca(OH)2 2, MgCl2 1, Mes 101, and Hepes 5; and was equilibrated to pH 7.6. This solution was altered for the ion selectivity experiments by replacing NaOH by KOH as appropriate to maintain a constant concentration of monovalent cations.
, 百拇医药
A Dagan TEV-200 amplifier or a NPI TurboTEC-10CX amplifier was used to record currents at 23°C in oocytes 2–3 days after injection with cRNA using standard two-electrode voltage-clamp techniques. Data acquisition was performed using a Pentium IV computer, a Digidata 1322 A/D interface and pClamp ver. 8 software (Axon Instruments). Specific pulse protocols used to voltage clamp oocytes are described in Results.
Data analysis
Analysis of data was performed with Clampfit 8 (Axon Instruments) and Origin 6 or 7 (Microcal) software. Briefly, the decaying phase of tail currents were fitted with a single exponential function and the fit was extrapolated to the beginning of the repolarization pulse. Extrapolated tail current amplitudes were normalized to the values determined after a pulse to +40 mV and the resulting data were fitted to a Boltzmann function:
, 百拇医药
where I is current amplitude, Imax is maximal tail current, Vt is the test voltage, V0.5 is the half-point of activation and k is the slope factor of the condcutance (G)–V relationship. G was calculated by measuring the current amplitude at the end of a 2-s pulse and dividing this value by the test potential minus the reversal potential (Vt–Erev) determined for each oocyte.
At –120 mV, WT KCNQ1 channels fully deactivated to a non-conducting closed state, whereas some of the mutant channels failed to close completely. To estimate the extent of channel closure at –120 mV, the tail currents elicited at this voltage after a step to 40 mV were fitted to a bi-exponential function. The slow component, representing deactivation, was extrapolated back to the beginning of the tail pulse and its amplitude at this point gave a value for the closing component a. In some cases mono-exponential fits were sufficient to describe deactivation because no hook in the tail was obvious. The difference between the steady-state non-deactivating component and zero current gave the value b. The calculated b/a ratios were used to estimate the relative channel open probability (Popen) at –120 mV.
, 百拇医药
Activating and deactivating current traces were fitted using a simplex algorithm to one (eqn (2)) or two (eqn (3)) exponential functions:
where y is current, A is current amplitude, t is time and is the time constant for the rise or decay of current.
Student's t test or ANOVA was used to test for statistical significance between groups. A value of P < 0.05 was considered statistically significant. Numerical values are reported as means ± S.E.M. (n = number of experiments).
, http://www.100md.com
Molecular modelling of the KCNQ1 channel pore
A 3-dimensional model of the S5 to S6 domains of the KCNQ1 subunit was constructed based on homology with the crystal structure of the corresponding domains of KcsA, a proton-activated K+ channel isolated from Streptomyces lividans (Doyle et al. 1998). The sequence of KCNQ1 is 46% homologous and 36% identical to KcsA within these regions. Initial models of KCNQ1 were built using Swiss-Model (Glaxo Welcome – http://www.expasy.org/swissmod/SWISS-MODEL.html) and energy minimized using GROMOS 96 (http://igc.ethz.ch/gromos) in the default configuration implemented in Swiss-Model.
, 百拇医药
Results
Kinetic analysis of WT and V310I KCNQ1 currents
Channels formed by homomultimeric assembly of subunits were characterized in oocytes injected with either WT KCNQ1 or V310I KCNQ1 cRNA. WT and mutant channels were expressed in oocytes and currents elicited by 2-s voltage-clamp pulses to potentials ranging from –100 to +60 mV from a holding potential of –70 mV. Both WT and V310I KCNQ1 channel currents activated with a multi-exponential time course in response to membrane depolarization and did not fully activate during a 2-s pulse (Fig. 1A and B). At potentials of +40 mV and higher the fast component of activation was followed by a transient reduction (due to faster rate of inactivation) and a subsequent second slower phase of activation. Tail currents were measured at –60 mV and were characterized by an initial increase in amplitude as channels recovered from inactivation, followed by a slower decay as channels deactivated.
, 百拇医药
A, WT KCNQ1 channel current elicited by 2-s pulses to potentials ranging from –100 to +60 mV, applied in 10 mV steps from a holding potential of –70 mV. Inset illustrates the voltage pulse protocol. B, V310I KCNQ1 channel currents elicited with the same protocol described for WT channels. For A and B, the calibration bars represent 1 s and 1 μA, respectively. C, relative conductance–voltage relationship for WT KCNQ1 channel currents () and V310I KCNQ1 currents () (n = 6–8). D, time constant of the fast activating component () plotted against test voltage for WT () and V310I () KCNQ1 currents (n = 5–8). E, time constants of deactivation () plotted against test voltage for WT () and V310I () KCNQ1 currents (n = 5–8). F, time constants for recovery from inactivation as a function of test potential for WT () and V310I () KCNQ1 currents (n = 5–8).
, 百拇医药
The voltage dependence of activation was determined from an analysis of tail currents elicited after 2-s activating pulses. The V0.5 was –21.7 mV and the slope factor was 13.9 mV for the voltage dependence of activation of WT KCNQ1 channels (n = 6). The V0.5 was –21.5 mV and the slope factor was 13.7 mV for V310I KCNQ1 channels (n = 6). The voltage dependence of WT and mutant KCNQ1 channel conductance was also estimated by measuring the amplitude of currents at the end of 2-s pulses applied to a variable potential, normalized to the driving force for K+ and plotted versus the test voltage (Fig. 1C). Peak V310I KCNQ1 conductance saturated near +60 mV, whereas the conductance of WT KCNQ1 progressively decreased at voltages above +20 mV, presumably due to increased channel inactivation. Although the conductance of V310I was altered, the kinetics of fast activation (Fig. 1D) was similar to WT channels. Deactivation was preceded by a transient increase in current, caused by rapid recovery of channels from inactivation in both WT and V310I channel currents. Tail currents elicited after a 2-s pulse to +40 mV were fitted to a bi-exponential function, where the fast and slow components reflected recovery from inactivation and subsequent deactivation, respectively. Neither measure of gating (recovery from inactivation or deactivation) was significantly altered by the V310I mutation (Fig. 1E and F).
, http://www.100md.com
Co-expression of WT and mutant KCNQ1 subunits with KCNE1
We recently reported that V310I currents are about 60% smaller than WT KCNQ1 currents and that IKs induced by co-expressing WT KCNE1 plus V310I KCNQ1 subunits was 89% smaller that WT IKs (Westenskow et al. 2004). Here, we extend the evaluation of V310I to include functional effects of co-expressing the different subunits in a ratio chosen to mimic the presumed expression pattern of an individual harbouring a heterozygous mutation. Oocytes were co-injected with cRNA encoding WT KCNQ1 (6 ng) and KCNE1 (0.15 ng), or WT KCNQ1 (3 ng) plus V310I KCNQ1 (3 ng) plus KCNE1 (0.15 ng) to determine functional effects of mutant subunits on IKs. KCNE1 cRNA (0.15 ng) was also injected alone to define the ‘background’ IKs which results from the interaction of exogenous KCNE1 and constitutively expressed KCNQ1 subunits (Sanguinetti et al. 1996). Examples of currents induced by injection of cRNAs are illustrated in Fig. 2A–C, and the current–voltage relationships for these experiments are summarized in Fig. 2D. The currents recorded in oocytes co-expressing KCNE1, WT KCNQ1 and V310I KCNQ1 were kinetically indistinguishable from oocytes expressing KCNE1 and WT KCNQ1 subunits alone. However, the magnitude of current was reduced by the presence of the mutant KCNQ1 subunit.
, 百拇医药
A–C, currents recorded from Xenopus oocytes injected with 6 ng WT KCNQ1 cRNA plus 0.15 ng KCNE1 cRNA (A), 0.15 ng KCNE1 cRNA alone (B) or 3 ng WT KCNQ1 cRNA plus 3 ng V310I KCNQ1 cRNA plus 0.15 ng KCNE1 cRNA (C). The scale bars in A represent 1 s and 1 μA. D, current–voltage relationships for currents measured at the end of 7-s pulses to the indicated test potentials.
Effects of Val310 mutations depend on volume of the substituted amino acid side group
, 百拇医药
The only obvious effects of the V310I mutation were to reduce the extent of inactivation of KCNQ1 and to reduce the amplitude of IKs. To explore further the potential influence of Val310 on the gating properties of KCNQ1 channels, this residue was mutated to Gly, Ala and Leu. These residues were chosen because of their similarity to Val and Ile with respect to hydrophobicity and molecular volume compared to the other natural amino acids. The mutant channels were expressed in oocytes and currents elicited by the same voltage pulse protocol described for WT channels in Fig. 1A. Representative traces of WT and mutant channel currents are shown in Fig. 3. The rate of current activation for the mutant channels was similar to WT channels. However, mutation of Val310 to smaller residues (Ala or Gly) greatly enhanced the rate of channel inactivation and prevented channels from fully closing (Fig. 3B and C). Tail currents were too complex to analyse kinetics, but they appeared to be hooked, as expected if channels recovered from inactivation before deactivation when the cell was repolarized to –60 mV. Channels containing a mutation of Val310 to a larger residue (Leu) activated and inactivated in a similar manner to WT or V310I KCNQ1, but the tail currents did not exhibit a hook. This does not necessarily imply that V310L channels have reduced inactivation compared to WT channels. Indeed, currents induced during depolarizing pulses to +30 mV exhibited a clear transient phase, suggesting channel inactivation. Instead, the lack of hooked tail currents suggests that recovery from inactivation was either too fast to observe, or very slow compared to the rate of deactivation. The normalized peak current–voltage relationship for V310L KCNQ1 was very similar to WT KCNQ1 channels (Fig. 4A). The V0.5 for the voltage dependence of activation for V310L KCNQ1 was –22.7 mV with a slope factor of 13.5 mV (n = 7). Because of extensive inactivation, the conductance–voltage relationships for V310A and V310G KCNQ1 (Fig. 4B) were determined from peak current values rather than from the end of the 2-s pulse. The current–voltage relationships were determined for both peak currents and currents at the end of 2-s test pulses (Fig. 4C and D). These current–voltage plots clearly show the extent of inactivation induced by the V310A and V310G mutations and that the channels did not close at potentials as negative as –100 mV. The ratio of current at the end of the pulse divided by the peak current for the test pulse to +40 mV was 0.14 for V310A and 0.26 for V310G KCNQ1. Thus, mutation of Val310 to smaller residues accentuated inactivation and appeared to prevent channel deactivation. However, as discussed below, V310G and V310A channels partially deactivated when tail currents were elicited at potentials more negative than –100 mV.
, http://www.100md.com
A, WT KCNQ1 currents and voltage pulse protocol. B–E, currents recorded from oocytes expressing the indicated Val310 mutant channels. In each panel, the scale bars represent 1 s and 1 μA.
A, conductance–voltage relationship for WT, V310I and V310L KCNQ1 (n = 6–7). B, peak currents plotted against the test pulse potential for V310G and V310A KCNQ1 channel currents (n = 6–8). C and D, peak currents for V310G and V310A KCNQ1 channels and currents measured at the end of 2-s test pulses to the indicated potentials (n = 6–8).
, 百拇医药
Val310 mutant channels deactivate faster than WT channels but fail to close normally
Tail currents for WT and mutant KCNQ1 channels were elicited by first pulsing to +40 mV to activate channels then returning the membrane to potentials ranging from –120 to –20 mV, applied in 20 mV increments. The rate of deactivation was slow for WT (Fig. 5A) and enhanced by mutations of Val310 (Fig. 5B–E). The fraction of channels that remained open at –120 mV (relative open probability, Popen) was calculated as described in Methods. Relative Popen was 0.02 ± 0.001 for WT KCNQ1 (n = 7) and 0.03 ± 0.001 (n = 7) for V310I KCNQ1. However, a small fraction of V310L channels remained open at –120 mV (Popen = 0.10 ± 0.01, n = 7), as did a substantial fraction of V310G channels (Popen = 0.32 ± 0.001, n = 6) and V310A channels (Popen = 0.48 ± 0.01, n = 13; Fig. 5F).
, 百拇医药
A–E, WT and mutant channel currents elicited by 2-s pulses, 125-ms pulses (V310G) or 300-ms pulses to +40 mV followed by pulses to different potentials as indicated by inset between A and B. F, relative open probability (Popen) determined at –120 mV for WT and mutant channels (n = 6–13).
Selective permeability of mutant channels
The relative cation selectivity of K+ channels is determined by the selectivity filter. Thus, based solely on the location of Val310, it is conceivable that mutations of this residue also could alter cation selectivity. To test for this possibility, the reversal potential (Erev) for KCNQ1 currents was determined for oocytes bathed in solutions containing an extracellular [K+] of 4, 16, 32 or 100 mM. Linear fits of the averaged data had slopes of 51–56 mV per 10-fold increase in [K+]o (Fig. 6), indicating that mutant channels maintained relatively normal K+ selectivity.
, http://www.100md.com
The reversal potential (Erev) for tail currents illustrated in Fig. 5 were measured and plotted against the extracellular [K+] for WT and mutant KCNQ1 channels expressed in oocytes. The slopes of the linear fits were 51–56 mV per 10-fold increase of extracellular [K+] (n = 6–8).
Residue Val310 putatively interacts with transmembrane helices S5 and S6
A 3-dimensional model of the pore domain of KCNQ1 was generated based on the high homology to the closed channel structure of the KcsA potassium channel (Fig. 7). The amino acid Val310 probably interacts with transmembrane helices S5 and S6, more specifically with Leu273 (S5), Phe332 (S6), Ala336 (S6) and Phe340 (S6). The residues were mutated and resulted in exchanges of the residue: Leu273 to Ala, Val, Phe or Gln, residue Phe332 to Ala, residue Ala336 to Gly or Cys and residue Phe340 to Ala, Ile or Tyr. These channels were expressed and L273A and L273Q were found to be non-functional. Further, F332A was very weakly functional. These channels could therefore not be studied and analysed in detail. Whereas all Phe340 mutant channels inactivated strongly the two functional Leu273 channels had a very variable impact on inactivation of channels (almost no inactivation in L273V and very strong inactivation in L273F). The tail currents of all other mutant channels were analysed and the relative open probability Popen was calculated. The Ala336 mutants closed normally. The mutant Leu273 and Phe344 channels could not close completely. However, the side chain volume at Leu273 did not matter (n = 6–8). The side chain volume of residue Phe340 determined Popen to some extend (Fig. 8). The larger residue Tyr at 340 allowed 83% (n = 8) of the channels to close whereas the smaller residues Ile and Ala allowed only 51% (n = 8) and 70% (n = 9) of channels, respectively, to close completely.
, 百拇医药
A KCNQ1 homology model was generated for the S5–S6 region based on the known KcsA crystal structure. The model shows residues Thr265–Tyr281 in the S5 domain in green and residues Trp323–Ile346 in the S6 domain in orange, all in a space-fill representation. The model was rotated 45 deg in the direction indicated by the arrows. The pore helix and the linkers are depicted as blue ribbons. Val310 is coloured red and residues which are in close proximity are labelled and coloured lighter than residues not in proximity to Val310.
, 百拇医药
A, representative current traces of mutant KCNQ1 channels. The current–voltage tracings were elicited with pulse protocols as used in Fig. 1. Mutant channel currents were further elicited by a 3-s pulse to +40 mV followed by pulses to potentials of –140 mV to –40 mV in 20 mV increments. The scale bars at the current–voltage traces represent 1 μA and 1 s. The current traces showing channel deactivation are scaled to about the same size to allow better view on the tail currents. B, relative open probability (Popen) determined at –120 mV for mutant channels (n = 5–9). n.f., non-functional. xb
, 百拇医药
Discussion
Most LQTS-associated missense mutations in KCNQ1 cause loss of function when the mutant subunits are expressed alone, and a dominant-negative effect when co-expressed with WT KCNQ1 and KCNE1 subunits (Shalaby et al. 1997). V310I is another example of a mutation that causes partial loss of function of IKs. Similar to previous studies (Barhanin et al. 1996; Sanguinetti et al. 1996), we found that current induced by injection of oocytes with KCNQ1 and KCNE1 cRNA was 25-fold larger than that induced by KCNE1 alone. In contrast, the average current induced by co-expression of V310I KCNQ1, WT KCNQ1 and KCNE1 subunits was 15-fold larger than that induced by KCNE1 alone. These data indicate that co-assembly of KCNQ1 and KCNE1 with V310I KCNQ1 subunits suppressed channel function, an effect that could result from a reduced single channel conductance, a higher flicker/block rate (Pusch et al. 2001), dysfunction of protein trafficking or protein misfolding and an accelerated rate of degradation. Whatever the exact mechanism, a decrease in repolarizing current would prolong action potential duration and QT interval, and increase the risk of ventricular arrhythmias.
, 百拇医药
The functional consequence of the V310I mutation was a simple loss of function. However, substitution of Val310 with smaller residues had a profound effect on channel gating. Mutation of Val310 to Gly or Ala greatly enhanced inactivation and prevented normal channel closure, whereas mutation to a larger residue (Leu or Ile) had only minor effects on channel gating. The relative hydrophobicity values (normalized to a value of zero for Gly) for the side chains of Ala, Val, Leu and Ile are –0.87, –3.10, –3.98 and –3.98. The accessible surface areas (ASA) for the same residues are 85, 113, 160, 180 and 182 2, respectively (Crighton, 1993). The median ASA for the residues in -helical proteins are 3, 5, 3, 4 and 4 suggesting that these hydrophobic residues are mostly hidden in the proteins (Lins et al. 2003). The gating of the V310G and V310A mutant channels were affected similarly, suggesting that size of the side chain was a factor contributing to the observed gating defects. KCNQ2–5 channels have a Leu in a position homologous to Val310 of KCNQ1. Mutation of Val310 to Leu in KCNQ1 eliminated the hooks normally observed in the tail currents, an effect that might be due to a faster rate of recovery from inactivation compared to channel deactivation.
, 百拇医药
Inactivation of voltage-gated K+ channels can occur by N-type (Hoshi et al. 1991), C-type (Lopez-Barneo et al. 1993) or P-type (De Biasi et al. 1993) mechanisms, named in reference to the structural region of the channel believed to participate in the gating mechanism. P refers to the pore domain and is often lumped together with C-type inactivation. It has been suggested that P-type inactivation involves a slight constriction of the narrow canal formed by the selectivity filter, causing a transient increase in the permeability of Na+ relative to K+ before the channel ceases to conduct (Kiss & Korn, 1998). C-type inactivation represents a stabilized P-type inactivated state of the channel (Loots & Isacoff, 1998). Perhaps reducing the volume of the side chain of residue 310 enhances the ability of the selectivity filter to constrict, leading to enhanced voltage-dependent inactivation. The location of residue 310 at the end of the pore helix closest to the selectivity filter limits the number of possible interactions with neighbouring residues. A homology model of the S5–S6 domains of KCNQ1 (Fig. 7), based on the structure of KcsA, reveals several residues that are in close proximity to Val310. Besides the immediately adjacent residues (Thr309 and Thr311), Val310 is in close contact with Leu273 of S5 and Ala336, Ile337 and Phe340 of S6. We mutated these residues and found that mutation of Leu273 to Val or Phe and Phe340 to small residues disrupted channel closure. Small residues would disrupt interaction of Phe340 side chain with Val310 based on modelling and together with the experimental data interaction of Phe340 with Val310 seems likely.
, 百拇医药
The LQTS-associated mutation L273F caused enhanced inactivation of KCNQ1 homomeric channels (Shalaby et al. 1997; Seebohm et al. 2001), an effect similar to that observed here for mutation of Val310 to Ala or Gly. Mutation of Leu273 to Val produced a weakly inactivating channel. Thus, the Val310 interaction with residue Leu273 does not only allow correct channel closure; it is also of high importance for channel inactivation. Perhaps a hydrophobic interaction between L273 of the S5 domain and Val310 of the pore helix stabilizes the open state of the selectivity filter (Liu & Joho, 1998). Disruption of hydrophobic interactions in the Val310/Leu273/Phe340 cluster by mutation of Val310 to the neutral amino acids Ala or Gly (or mutation of Leu273 to Phe or Phe340 to small or hydrophilic amino acids) might destabilize the open state of the pore, causing an enhanced rate and extent of inactivation. Disruption of these interactions by V310I and L273F might cause LQTS associated with KCNQ1 mutant channels. It seems plausible that nearby mutations also influence the Val310–Leu273–Phe340 interactions. In fact, every single amino acid in the lower pore helix (305–312) and several in S5 (266, 269, 273, 275–277) and S6 (341, 342, 344, 345) result in reduced function (summary: http://pc4.fsm.it:81/cardmoc/) and LQTS.
, http://www.100md.com
Given the suspected importance of the pore helix and surrounding regions in the process of P- and C-type inactivation, perhaps it was not too surprising to discover that mutations of Val310 to other amino acids could affect inactivation. However, the finding that mutation of this residue to Gly or Ala impaired channel closure was totally unexpected. The mechanism of incomplete deactivation in these mutant channels is unknown, but may relate to the relative ability of the S6 domain to bend back and forth at the proposed hinge point for the activation gate. The hinge for the activation gate of the bacterial K+ channel MthK is a Gly residue located near the midpoint of the S6 domain (Jiang et al. 2002). This Gly is conserved in most, but not all Kv channels. In KCNQ1, this residue is an Ala (Ala336), located close to residue 310 (Fig. 7). Mutation of Val310 to a smaller residue (Gly or Ala) may not prevent the full range of bending and/or twisting of S6 required for normal closure of the activation gate. Indeed, we have studied the impact of the Gly at the putative glycine hinge (KCNQ1 Ala336) and found it to be of minor importance for KCNQ1 channel gating (authors' unpublished observations). It is more likely that the lack of channel closure might be due to disruption of a pore helix–S6 interaction similar to that previously described for Kv2.1 channels. It was proposed that interaction of a residue near the base of the pore helix (Thr370) with a S6 residue (Cys393) in Kv2.1 stabilizes the open state and that counter-clockwise rotation of S6 at negative transmembrane potentials disrupts this interaction, allowing the channel to close (Liu & Joho, 1998). Val310 of KCNQ1 is located next to the residue homologous to Thr370 of Kv2.1, and Phe332 of KCNQ1 is in the same position as C393 of Kv2.1. Mutation of a single S6 residue (Cys480) in exp-2 channels of C. elegans has also been associated with disrupted deactivation. This affect was attributed to inability of the mutant S6 domain to rotate normally during gating due to a steric hindrance of the bulky aromatic group with specific residues in the adjoining S5 domain (Liu & Joho, 1998; Espinosa et al. 2001). Because KCNQ1 is not strictly dependent on a gating hinge (authors' unpublished observations) the proposed gating of KCNQ1 might include straighter S5 and S6 helices compared to the pronounced gating hinge in MthK. It seems possible that the central important positioning of the pore helix, S5 and S6 have a strong impact on channel state stability.
, http://www.100md.com
In summary, mutation of Val310, a residue located at the base of the pore helix of KCNQ1, can alter channel gating associated with inactivation and deactivation. Homology modelling and experimental data suggest a complex interaction between residue 310 and closely situated residues in the pore helix, S5 (Leu273) and S6 (Phe340). These interactions may determine the stability of two gates, one associated with channel activation and deactivation and formed by criss-crossing of the intracellular ends of the S6 domains (Doyle et al. 1998), and another one associated with P/C-type inactivation and formed by the selectivity filter–pore helix complex (Loots & Isacoff, 1998). Disruption of these interactions might underly LQTS associated with KCNQ1 mutant channels V310I and L273F and possibly channels with nearby mutations.
, 百拇医药
References
Barhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M & Romey G (1996). K(V)LQT1 and lsK (minK) proteins associate to form the I Ks cardiac potassium current. Nature 384, 78–80.
Crighton TE (1993). Proteins Structures and Molecular Properties 2nd edn. HW Freeman, New York.
De Biasi M, Drewe JA, Kirsch GE & Brown AM (1993). Histidine substitution identifies a surface position and confers Cs+ selectivity on a K+ pore. Biophys J 65, 1235–1242.
, 百拇医药
Doyle DA, Morais CJ, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT & MacKinnon R (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69–77.
Espinosa F, Fleischhauer R, McMahon A & Joho RH (2001). Dynamic interaction of S5 and S6 during voltage-controlled gating in a potassium channel. J Gen Physiol 118, 157–170.
Hoshi T, Zagotta WN & Aldrich RW (1991). Two types of inactivation in Shaker K+ channels: effects of alterations in the carboxy-terminal region. Neuron 7, 547–556.
, 百拇医药
Jiang Y, Lee A, Chen J, Cadene M, Chait BT & MacKinnon R (2002). The open pore conformation of potassium channels. Nature 417, 523–526.
Kiss L & Korn SJ (1998). Modulation of C-type inactivation by K+ at the potassium channel selectivity filter. Biophys J 74, 1840–1849.
Lins L, Thomas A & Brasseur R (2003). Analysis of accessible surface of residues in proteins. Protein Sci 12, 1406–1417.
, 百拇医药
Liu Y & Joho RH (1998). A side chain in S6 influences both open-state stability and ion permeation in a voltage-gated K+ channel. Pflugers Arch 435, 654–661.
Loots E & Isacoff EY (1998). Protein rearrangements underlying slow inactivation of the Shaker K+ channel. J Gen Physiol 112, 377–389.
Lopez-Barneo J, Hoshi T, Heinemann SH & Aldrich RW (1993). Effects of external cations and mutations in the pore region on C-type inactivation of Shaker potassium channels. Receptors Channels 1, 61–71.
, http://www.100md.com
Melman YF, Um SY, Krumerman A, Kagan A & McDonald TV (2004). KCNE1 binds to the KCNQ1 pore to regulate potassium channel activity. Neuron 42, 927–937.
Morais-Cabral JH, Zhou Y & MacKinnon R (2001). Energetic optimization of ion conduction rate by the K+ selectivity filter. Nature 414, 37–42.
Pusch M (1998). Increase of the single-channel conductance of KvLQT1 potassium channels induced by the association with minK. Pflugers Arch 437, 172–174.
, 百拇医药
Pusch M, Ferrera L & Friedrich T (2001). Two open states and rate-limiting gating steps revealed by intracellular Na+ block of human KCNQ1 and KCNQ1/KCNE1 K+ channels. J Physiol 533, 135–143.
Romey G, Attali B, Chouabe C, Abitbol I, Guillemare E, Barhanin J & Lazdunski M (1997). Molecular mechanism and functional significance of the MinK control of the KvLQT1 channel activity. J Biol Chem 272, 16713–16716.
Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS, Atkinson DL & Keating MT (1996). Coassembly of K(V)LQT1 and minK (IsK) proteins to form cardiac IKs potassium channel. Nature 384, 80–83.
, 百拇医药
Sarkar G & Sommer SS (1990). The ‘megaprimer’ method of site-directed mutagenesis. Biotechniques 8, 404–407.
Seebohm G, Scherer CR, Busch AE & Lerche C (2001). Identification of specific pore residues mediating KCNQ1 inactivation. A novel mechanism for long QT syndrome. J Biol Chem 276, 13600–13605.
Sesti F & Goldstein SA (1998). Single-channel characteristics of wild-type IKs channels and channels formed with two minK mutants that cause long QT syndrome. J Gen Physiol 112, 651–663.
, http://www.100md.com
Shalaby FY, Levesque PC, Yang WP, Little WA, Conder ML, Jenkins-West T & Blanar MA (1997). Dominant-negative KvLQT1 mutations underlie the LQT1 form of long QT syndrome. Circulation 96, 1733–1736.
Smith PL, Baukrowitz T & Yellen G (1996). The inward rectification mechanism of the HERG cardiac potassium channel. Nature 379, 833–836.
Tai KK & Goldstein SA (1998). The conduction pore of a cardiac potassium channel. Nature 391, 605–608.
, 百拇医药
Westenskow P, Splawski I, Timothy KW, Keating MT & Sanguinetti MC (2004). Compound mutations: a common cause of severe long-QT syndrome. Circulation 109, 1834–1841.
Yang Y & Sigworth FJ (1998). Single-channel properties of IKs potassium channels. J Gen Physiol 112, 665–678.
Zhou Y, Morais-Cabral JH, Kaufman A & MacKinnon R (2001). Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution. Nature 414, 43–48., http://www.100md.com(Guiscard Seebohm, Peter W)