|
|
|
|
|
|
|
||
Electrostatic Control and Chloride Regulation of the Fast Gating of ClC-0 Chloride Channels
Address correspondence to Tsung-Yu Chen, Center for Neuroscience University of California-Davis, 1544 Newton Court, Davis, CA 95616. Fax: (530) 754-5036; email: tycchen{at}ucdavis.edu
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: ClC gating • electrostatic effect • foot-in-the-door
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Jentsch et al., 1990; O'Neill et al., 1991) is the prototype of the ClC channel family, and its functional properties are the best known among all ClC members (for review see Jentsch et al., 1999, 2002; Maduke et al., 2000). This channel is a two-pore homodimer (Middleton et al., 1994, 1996; Ludewig et al., 1996; Lin and Chen, 2000), and the opening and closing of the channel pores are controlled by two functionally distinct gating mechanisms: the fast and the slow (inactivation) gating (Miller, 1982; Miller and White, 1984; Richard and Miller, 1990). The fast gating of ClC-0 controls each individual pore with kinetics in the millisecond range. On the other hand, the slow gating occurs in the time range of seconds, and this slower mechanism opens and closes the two pores simultaneously (for reviews see Miller and Richards, 1990; Maduke et al., 2000). These two gating mechanisms show an interesting feature: the gating appears to couple to the movement of Cl- in the pore. Consequently, the gating properties are controlled by the Cl- concentrations ([Cl-]) in the extracellular and intracellular solutions (Richard and Miller, 1990; Pusch et al., 1995; Chen and Miller, 1996).
At the phenomenological level, changing both the external ([Cl-]o) and internal Cl- concentrations ([Cl-]i) alters the fast-gating properties of the channel. External Cl- is crucial in opening the fast gate. When [Cl-]o is raised, the opening rate of the fast gate is increased, and this contributes to the apparent voltage dependence of the fast gating (Pusch et al., 1995; Chen and Miller, 1996). A kinetic model has been proposed to explain how the external Cl- can open the fast gate in a depolarization-activated manner (Chen and Miller, 1996). Internal Cl- also regulates the fast gating. The effect, however, is more prominent on the closing rate of the fast gate (Chen and Miller, 1996). It has been qualitatively described that when [Cl-]i is elevated, the closing rate of the fast gate is reduced, a phenomenon very similar to the "foot-in-the-door" effect previously described in cation channels (Swenson and Armstrong, 1981).
Understanding the control of the ClC-0 fast gating by Cl- ions has been greatly facilitated by the recently solved high-resolution X-ray structures of the bacterial ClC channels (Dutzler et al., 2002, 2003). Fig. 1 shows a side-view of the inner pore region of the ClC channel with several residues likely lining the pore. The structures from bacterial ClC channels show that a glutamate residue, E148 (the residue in black on top of Fig. 1), which corresponds to E166 in ClC-0, projects its side chain into the ion permeation pathway. This glutamate side chain has been proposed to be the channel gate (Dutzler et al., 2002). In addition, three Cl--binding sites were identified in the pore of mutant forms of the channel, in which E148 is mutated to alanine (E148A) or glutamine (E148Q) (Dutzler et al., 2003). The first binding site is at the center of the pore (Scen), and the bound Cl- is stabilized by the dipoles of helices and the sidechain hydroxyl groups of the pore residues. A second Cl--binding site is internal to Scen, and is located at a position where the intracellular aqueous vestibule meets the selectivity filter (Sint). Finally, in the E148 mutants, a third anion (not depicted) is observed in place of the negative charge on the E148 side chain in the wild-type (WT) channel. This Cl--binding site, which is named Sext, is only 4 Å external to Scen. The presence of Cl- at Sext when the negative charge on the glutamate side chain is removed by mutation suggests a plausible mechanism for the gating-permeation coupling—a competition of Cl- with the side chain of E166 (Dutzler et al., 2002, 2003).
|
). K519 of ClC-0 is located at the intracellular pore entrance and is 
20 Å away from E166 based on the corresponding distance in the bacterial channel structure (Dutzler et al., 2002). If E166 is the fast gate, how could the mutation of K519 affect the fast gating? Even at the phenomenological level, it is not known if the mutation of K519 affects the opening rate, the closing rate, or both, since these two kinetic parameters can be differentially regulated by Cl- ions coming into the pore from different directions.
The residue K519 of ClC-0 is important not only in the fast-gating mechanism but also in the control of channel conductance. We have recently shown that the charge on this residue contributes to the intrinsic electrostatic potential of the pore, which determines the conductance of the channel (Chen and Chen, 2003), and, in the case of pore cysteine mutants, controls the cysteine modification rate by charged methane thiosulfonate (MTS) reagents (Lin and Chen, 2003). In addition, we also showed that mutations of other residues lining the inner pore region change the ion flux across the pore. The positions of these residues, including E127, Y512, I515, and K519, and their relations to the proposed glutamate gate can be viewed from the stereo picture in Fig. 1. Since the gating of the channel is thought to be tightly linked to ion permeation, we suspect that the mutations of these pore-lining residues would also affect the fast gating of the channel via the permeant ions in the pore. In this study, we examine the effects of [Cl-]i and the intrinsic electrostatic potential of the pore on the fast-gating to explore the mechanistic operation of ClC-0. We trace the path of a Cl- ion and analyze the gating parameters from mutants that likely alter the Cl- occupancy in the pore. The results reveal that internal Cl- exerts an effect on the closing rate of the fast gate with a much lower apparent affinity than that of the external Cl- effect on the opening rate. When the pore residues are made more positive by mutations, the opening and the closing rates of the fast gate are increased. On the other hand, when the intrinsic electrostatic potential of the pore is more negative, the fast gating is slower. Thus, the effect of [Cl-]i on the fast-gate closing rate may come not only from a direct competition of Cl- with the negatively charged glutamate side chain, as suggested from the X-ray crystal structures of the bacterial ClC channel, but also from an overall more negative potential brought into the pore by Cl- binding. Because the opening and the closing of the ClC-0 fast gate are controlled by external and internal Cl- in a very different way, we also suggest that these two kinetic aspects of the fast gating likely are not the two opposite directions of a reversible gating process, but form a gating cycle during the operation of the ClC-0 fast gating.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Most of the mutations were constructed in the background of the C212S mutation to remove the inactivation of the channel (Lin et al., 1999). Some of the mutants were also constructed in the WT background, and the difference in fast gating in the WT and C212S background is not significant. We therefore refer to the C212S mutant as the WT channel. All the channels examined were expressed in Xenopus oocytes, and RNA synthesis and oocyte preparation/injection were the same as those described previously (Chen, 1998; Lin et al., 1999).
Single-channel Recordings
Excised inside-out patch configuration (Hamill et al., 1981) was performed in all single-channel recordings. The recording pipettes were pulled from borosilicate glass capillaries by a pipette puller PP-830 (Narashige), and had resistance of 3–6 M
when filled with the pipette solution. In all experiments, the pipette (external) solution was the same and contained (in mM): 110 NMDG-Cl, 5 MgCl2, 1 CaCl2, 5 HEPES, pH 7.5. The standard bath (internal) solution in which high G seals were made contained (in mM): 110 NaCl, 5 MgCl2, 1 EGTA, 5 HEPES, pH 7.5. Various internal solutions containing 30–2,400 mM [Cl-]i have a content of (in mM): (X-10) NaCl, 5 MgCl2, 1 EGTA, 5 HEPES, pH 7.5, where X represents the indicated [Cl-]i. For solutions containing [Cl-]i <120 mM (such as 30 or 60 mM), Na-glutamate was added to make the ionic strength similar to the 120 mM Cl- solution. The pH's of the internal and external solutions were adjusted with NaOH and NMDG, respectively. For most single-channel recordings, the current was online filtered at 0.2 kHz (-3 dB), and digitized at 1 kHz. In analysis the current was further digitally filtered at 0.2 kHz, leading to a final cut-off frequency at 140 Hz. For mutants with fast kinetics (such as I515K and D513S, see Fig. 7), the filter frequency and the sampling rate were 0.5 and 2.5 kHz respectively. The exchange of internal solution was achieved using a SF-77 solution exchange system (Warner Instruments, Inc.) as described previously (Chen and Chen, 2003). The presented voltages in the current study were not corrected for junction potentials.
Data Analysis
Because some mutants were constructed in the WT background, the inactivation events of the channel were sometimes difficult to separate from the closed events. For all the results presented, the analysis of the single-channel traces, which comprise three equidistant current levels, was made according to the method described previously (Chen and Miller, 1996; Chen and Chen 2001). This analysis method exploits the fact that the ratio of the state-probabilities for the fully-open and the middle-current level (f2/f1) as well as the dwell-time of the events at these two current levels (
2 and 1, respectively) are completely independent of the inactivation time. Thus, under the assumption that the channel contains two pores independently controlled by their fast gates (Miller, 1982; Hanke and Miller, 1983; Miller and White, 1984; Bauer et al., 1991; Middleton et al., 1996; Ludewig et al., 1996; Lin and Chen, 2000), the overall open probability (Po) of the fast gate is determined from eq. 1:
 | (1) |
= f2/f1. The opening rate (
) and the closing rate (ß) are calculated according to Eqs. 2a and b:
 | (2a) |
 | (2b) |
To show the saturation effect on the fast gating of the channel by increasing [Cl-]i, we plotted the averaged open duration of an individual pore (o) as a function of internal Cl- activity. For comparison, the averaged duration of the closed events (c) was also estimated from the opening rate. The estimate of o and c was based on Eqs. 3a and b:
 | (3a) |
 | (3b) |
The dependence of the open duration of the fast gate on the internal Cl- activity is fitted by an empirical hyperbolic equation:
 | (4) |
min and max are the open duration of the channel at zero and infinite internal Cl- activity, respectively, ACl is the internal Cl- activity, and K1/2 is the Cl- activity at which the half maximal effect on the open duration was observed. The calculation of the internal Cl- activity was based on the equation ACl = µ[Cl-], using the following activity coefficients (µ): 30 mM, 0.930; 60 mM, 0.867; 120 mM, 0.769; 300 mM, 0.710; 600 mM, 0.673; 1,200 mM, 0.654; 2,400 mM, 0.684 (Robinson and Stokes, 1955). Throughout the whole paper, the averaged data were expressed as mean ± SEM. Curve fitting was performed with an un-weighted, least-squares method using Origin software (OriginLab Co.). |
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). To quantitatively characterize the internal Cl- effect, we recorded single ClC-0 channels with the excised inside-out configuration, and systematically varied [Cl-]i from 30–2,400 mM. Fig. 2 A shows the representative single-channel recording traces of the WT channels and their analyses with dwell-time histograms of the three current levels. While raising [Cl-]i from 120 to 2,400 mM has little effect on the durations of closed events (represented by open squares in the dwell-time histograms), the same increases in [Cl-]i significantly prolong the durations of the fully open events (filled circles in the dwell-time histograms). As the durations of the closed and open events are inverse functions of the opening and closing rates, respectively (Eqs. 3a and b), the dwell-time histograms demonstrate that the major effect of [Cl-]i is on the closing rate of the fast gate. Fig. 2 B shows the averaged opening and closing rates of the channel at various [Cl-]i as a function of membrane voltage. Although an increase of [Cl-]i appears to slightly reduce the opening rate at the depolarized potentials, the effect is very small (Fig. 2 B, left). On the other hand, increasing [Cl-]i significantly reduces the closing rate so that the entire closing rate-voltage curve is shifted in parallel (Fig. 2 B, right). These results are consistent with those in the previous single-channel study using native ClC-0 channels directly purified from Torpedo electric organ (Chen and Miller, 1996).
|
first drew attention to a basic residue K519 in ClC-0. By mutating this positively charged residue to glutamate (K519E), they showed that the fast gating of the channel was altered. However, since K519 was located on the intracellular side of the membrane, these authors could not conclude that K519 was directly involved in forming the anion binding site mediating the external Cl- effect (Pusch et al., 1995). Thus, how the mutation of this residue affected the channel gating was not known. To investigate the functional role of K519 in the Cl- regulation of fast gating, we compare the fast-gating kinetics among the WT (K519), K519C, and K519E channels at the single-channel level (Fig. 3). Representative single-channel traces and dwell-time histograms are shown in Fig. 3 A, and the averaged results of the opening and closing rates at [Cl-]i = 600 mM are plotted as a function of the membrane voltage in Fig. 3 B. It can be seen that mutation of K519, again, alters the closing rate much more than the opening rate. Furthermore, when the side chain charge of residue 519 is made more negative, the closing rate is reduced, and the closing rate-voltage curve is also shifted in parallel, a phenomenon similar to that of raising [Cl-]i. As for the WT channel, varying [Cl-]i has little effect on the opening rate in the mutant channels. On the other hand, the closing rates of these channels are affected significantly by altering [Cl-]i (see below). This "foot-in-the-door"-like effect (Swenson and Armstrong, 1981)—the higher the [Cl-]i, the longer the duration of the open events—can also be directly observed from previous published recording traces (see Fig. 4 of Chen and Chen, 2003).
|
|
c, reflecting the opening rate) and the open durations (o, reflecting the closing rate) of the fast gate of these three channels are directly compared in Fig. 4, A and B, respectively. In this figure, we calculate c and o from the recording traces at -110 mV, and plot the data as a function of internal Cl- activity. The values of c for these three channels are similar to each other, ranging from 15–35 ms, and raising [Cl-]i from 30 to 2,400 mM (or ACl from 28 to 1642 mM) does not significantly change the value of c (Fig. 4 A). On the other hand, a very similar pattern of the internal Cl- effect on o is observed for all three channels—the open duration increases with Cl- activity, but the relation is not a straight line (Fig. 4 B). This result is consistent with the idea that the effect may be mediated through a Cl--binding site. For the WT channel, fitting the data points to an empirical hyperbolic relation shows a half-saturating Cl- activity of 1,300 mM (Fig. 4 B), a value significantly higher than the K1/2 in the Cl- titration curve for the pore conductance (60–75 mM, see White and Miller, 1981; Chen and Chen, 2003). It appears that there is a finite open duration of 3–4 ms even in the absence of intracellular Cl-. For the K519C and K519E mutants, the Cl- titration curves are shifted in parallel toward longer open duration, and the more negative the charge at position 519, the longer is the open duration of the fast gate. Therefore the charge at position 519 controls the minimal open duration in the absence of internal Cl- but appears to have less effect on the apparent affinity of the potential Cl--binding site that is responsible for this effect.
Because the closing rate is affected by Cl- ions and the charge mutation in the pore in a similar manner (Figs. 2 and 3), we ask if the alteration in the closing rate by Cl- is a specific effect to the permeant ions. To this end, we examine if other anions, such as bromide (Br-), a permeant ion, or sulfate (SO42-), a nonpermeant ion, can exert similar gating effects. In these experiments, [Cl-]i was constant in the intracellular solutions (300 mM in Br- experiments and 120 mM in SO42- experiments), and various concentrations of Br- or SO42- were added to the solutions. Fig. 5 A shows the single-channel conductance of the WT channel in the presence of various concentrations of internal Br-. The conductance of the channel is reduced by Br-, with an estimated apparent blocking affinity of 24 mM (Fig. 5 A). Because Br- can permeate through the pore (unpublished data), the half-blocking concentration here may not precisely reflect the true blocking affinity of Br-. From the same experiments, the open durations of the fast gate are estimated and shown in Fig. 5 B. Similar to the internal Cl- effect on the channel gating, an elevation of [Br-]i increases the open duration of the fast gate. The titration curve for this Br- effect is not linear, similar to the action of Cl-. The effect of [Br-]i on the open duration of the channel is stronger than that exerted by [Cl-]i because at the same total concentration of the internal anions the open duration is longer in the presence of Br-. On the other hand, the nonpermeant ion SO42- does not have an appreciable effect on the open duration of the fast gate. In the presence of 120 mM [Cl-]i with additional 180 or 480 mM SO42-, the open duration is the same as that in the absence of this nonpermeant ion (Fig. 5 B).
|
). However, the functional role of this mutation on the fast gating is revealed in the concurrent mutations of E127 and K519. After removing the negative charge at position 127, the channels with either a positive, neutral, or negative charge at position 519 all have a similar open duration of the fast gate (Fig. 6 A, closed circles in dwell-time histograms). The slight differences in the duration of the closed events (open squares) among the three dwell-time histograms appear to be within a normal variation from patch to patch. Fig. 6, B and C, compare the averaged closed and open durations of the three K519 mutations in the absence and in the presence of the E127Q mutation. The closed durations (reflecting opening rate) from the three channels are 15–35 ms at all [Cl-]i, and the effects from charge alterations are absent whether E127Q mutation is present or not, indicating that E127Q mutation has limited influences on the fast-gate opening mechanism (Fig. 6 B). On the other hand, even though the open duration in these mutants is still controlled by internal Cl- to the same degree as that in the WT channel, the open durations of the three K519 mutants in E127Q background are almost identical to each other, and they are also similar to those of the WT channel at various [Cl-]i (Fig. 6 C). These results indicate that E127 residue did play an important role in regulating the closing rate of the fast gate. However, it appears that the negative charge of this residue interacts with the positive charge from the K519 side chain in an unknown manner to exert the electrostatic control on the fast-gating mechanism.
|
2.5- and 7.3-fold, whereas removing a negative charge at position 513 (WT to D513S) increases the opening and closing rate by 4.0- and 6.7-fold, respectively (Fig. 7 B). We also replaced E127 with a positively charged residue. The single-point mutant E127K was not functionally expressed. We therefore compared the gating behavior of the double mutant E127K/K519E with that of E127Q/K519E (Fig. 7 C). Again, a more positive charge at position 127 also increases the closing rate of the channel by 50% even though the opening rate in both mutant channels is not significantly different from that of the WT channel (Fig. 7 D). The results thus demonstrate that introducing positively charged residues (or removing negatively charged residues) in the pore consistently speeds up the fast gating of ClC-0. On the other hand, introducing negative charges in the pore (for example, K519E and I515E/E127Q mutants, or K519C modified with 2-sulfonatoethyl methanethiosulfonate (MTSES), unpublished data) makes the fast gating become slower. Thus, increasing [Cl-]i and introducing more negative charge to the pore lead to a similar effect—an energetic stabilization of the open state of the fast gate.
|
). The conductance of the mutant S123T was consistently small throughout the whole range of [Cl-]i, and the duration of the open events appeared longer than the WT channel as judged from the single-channel traces (Fig. 8 A, lower traces). The small conductance of this mutant, however, prevents an accurate evaluation of the gating parameters. Consequently, we only focused on the analysis of the gating parameters of the Y512F mutant. The durations of the fully open events of the fast gate of Y512F appear to be longer than those of the WT channel at both 300 and 2,400 mM [Cl-]i (Fig. 8 A, middle traces). Fig. 8, B and C, compare the averaged closed (c, the inverse of the opening rate) and open durations (o, the inverse of the closing rate) of Y512F with those of the charged mutants at position 519. It is apparent that for all of these channels, varying [Cl-]i significantly changes the o (or the closing rate) but not c (opening rate). The o of the Y512F mutant approaches that of the WT channel at low [Cl-]i. At high [Cl-]i, however, the o of the Y512F mutant becomes even longer than that of the K519E mutant. Curve fitting for the data points reveals that the saturated open duration of the Y512F mutant is approximately twofold longer than that of the WT channel. The apparent affinity of Cl- in mediating the effect, however, is roughly similar to that of the WT channel. Thus, the mutation Y512F does not change the opening rate of the fast gate, but alters the range of the fast-gate closing rate that is controlled by [Cl-]i.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The analogy of "foot-in-the-door" (Chen and Miller, 1996) implies that the effect is mediated by a binding site (or binding sites) close to the channel gate. When examining the dependence of the open duration on the internal Cl- activity, we found that the relation is not linear, but is described by a hyperbolic curve. This observation is significant because it implies that the Cl- effect on the open duration is saturable. The fact that Br- but not SO42- can exert a similar gating effect (Fig. 5 B) shows that permeant ions can reach this binding site but not nonpermeant ions. Finally, the effect of Br- in reducing the closing rate appears to be larger than that of Cl- (Fig. 5 B). This may be due to a higher Br--binding affinity than that of Cl- in the pore. These results are all consistent with the idea that the Cl- regulation on the closing rate is through specific ion-binding sites in the pore. However, the Cl- titration curve reveals that Cl- has a very low apparent affinity in mediating this gating effect, with a half-saturating Cl- activity in the molar range. In comparison, the half-saturating Cl- activity for the channel conductance is 60–75 mM (Chen and Chen, 2003; also see White and Miller, 1981). The results from Br- experiments also suggest that the affinity of Br- in increasing the open duration is significantly lower than that of Br- in reducing the channel conductance (Fig. 5). Therefore, although controls of channel gating and pore conductance are likely to be mediated by the same ion-binding sites in the pore, the contribution of each binding site to the regulation of the fast-gate closing and to the determination of the channel conductance is probably different.
In addition to the Cl- effect on the open duration, we have found that charge mutations in the inner pore region also alter the close-open transition rate of the fast gating. Although the mutations could create unexpected conformational changes that might allosterically alter the fast gating, for several reasons we think that a major portion of the mutation effects on the fast-gate closure comes from an electrostatic mechanism. First, the effects of mutations at position 519 concur with the charge instead of the size or shape of the mutated side chain. Thus, the fast-gating of K519F or K519M is similar to that of K519C while the gating of K519D is similar to K519E (see recording traces in Fig. 8 of Chen and Chen, 2003). Second, the effect on the fast-gate closure (duration of open events) from the mutation of a positively charged residue K519 can be nearly eliminated by a simultaneous counter-charge mutation, E127Q (Fig. 6). Finally, we have consistently found that increasing the positive charge in the inner pore region increases the open-close transition rate of the fast gate (Fig. 7). The effect is most notable when charge manipulations occurred at I515 and D513. Although the effect is less prominent at position 127, we suspect that this is likely due to the fact that E127 may interact with K519, and therefore its negativity is reduced by the nearby positive charge from K519. Likewise, when the double mutant E127K/K519E was made, the positive potential of E127K was reduced by the negative charge of the K519E side chain. However, comparing the E127K/K519E mutant with the E127Q/K519E mutant still shows that the former double mutant has a faster closing rate than the latter (Fig. 7, C and D). Therefore, the results from charge mutations at position 127 and 519 are also consistent with those results obtained at positions 515 and 513—a more positive potential in the pore results in a faster fast gating.
The observation that changing [Cl-]i and altering electrostatic potential in the pore both affect the fast gating in a similar manner—for example, influence the closing rate more than the opening rate, and shift the closing rate curve in parallel—suggests that these two different manipulations may act through similar mechanisms. What are the possible mechanisms underlying these effects? The crystal structures of the bacterial ClC channels reveal that the side chain of E148 occludes the ion permeation pathway, suggesting that the side chain of the corresponding glutamate (E166) in ClC-0 might be the gate (Dutzler et al., 2002). The structures of the E148 mutants of the bacterial channel further show that the pore contains an uninterrupted queue of three Cl- ions in the pore (Dutzler et al., 2003). In particular, one of the Cl- ions binds to the external Cl- binding site, Sext, in place of the negatively charged side chain of E148 in the wild-type bacterial channel. This structural picture suggests that Cl- may occupy Sext position to compete with the negatively charged side chain of E166 of ClC-0, and thus prevent the E166 side chain from adopting a closed position (Dutzler et al., 2003). Such a competition mechanism, therefore, almost perfectly explains the "foot-in-the-door" effect exerted by the internal Cl-.
Under the assumption that E166 of ClC-0 is the fast gate, we considered two possibilities for how the charge mutations at the inner pore region could influence the fast gating. First, the negative charge on the gate could directly interact, via a through-space electrostatic force, with the charged residues in the inner pore region. A more negative potential from the inner pore region would repel the gate so that it is more difficult to close. Although some of the charge mutations shown above are fairly distant from the side chain of E166—for example, the charge on the side chains of E127 and K519 are 15 and 20 Å away from the side chain of E166, respectively—the dielectric constant within protein molecules is usually much smaller than that in the bulk water solution. A direct electrostatic influence on the glutamate gate, however, cannot completely explain all the phenomena. For example, the WT channel (E127/K519) and the E127K/K519E double mutant have their positive and negative charges swapped at positions 127 and 519, but the charge pair is preserved. However, the fast-gate closing rate was increased by 50% in the double mutant (Fig. 7 D). One may argue that since position 127 is deeper in the pore than position 519, it is closer to the negatively charged residue E166. Consequently, E127K/K519E channel would exert a more positive potential toward the glutamate gate, leading to a faster gating mechanism. However, if a positive charge at position 127 can have an electrostatic influence on the glutamate gate, why would a removal of the negative charge at this position (E127Q) have little effect on the fast-gate closure?
A second possibility to explain the effect of charge mutation on the fast gating would invoke a mediating role for the permeant ion. Mutation of E127Q by itself has no effect on the gating. However, this mutation nearly removes the mutational effects of K519 on the closing rate. Although the nature of the asymmetric effects from the mutations of these two neighboring residues is unknown, the influence on gating by the E127Q mutation mirrors the effect of this mutation on channel conductance (Chen and Chen, 2003). Such a similarity argues that the mutational effect on the fast gating could be partly mediated by permeant ions. For example, the charge effect on the fast gating may involve the electrostatic force in the pore from the Cl- at Sint to that at Sext. The pore conductance is controlled by the charge placed at positions 127, 515, and 519, which are located on the intracellular side of Sint (Chen and Chen, 2003). When more positive charges are placed at these positions, the potential for Cl- at Sint to repel the Cl- at Scen and then the one at Sext would be decreased. This would decrease the competition of Cl- for Sext with the side chain of E166. Consequently, the E166 side chain may more easily assume its closed position—a faster closing rate of the fast gate.
Although this ion–ion interaction mechanism qualitatively explains the electrostatic control on the fast gating, a quantitative comparison between the effects on pore conductance and fast gating is not completely satisfactory. For example, the E127K/K519E double mutant has a similar conductance-Cl- activity curve as that of the I515K mutant (Fig. 9 in Chen and Chen, 2003). However, the fast gating of the latter is much faster than the former. Similarly, the D513S and I515K mutants have a similar closing rate, but their conductances are quite different (Fig. 7, A and B). We suspect that there is a position effect, and depending on the location where the charge is manipulated, the above two mechanisms may have different contributions. At positions more distant from the permeation pathway (for example D513), a through-space electrostatic effect may be more important. On the other hand, at positions closer to permeant ions, like E127 and K519, the electrostatic effect mediated by the permeant ions may be more prominent. Since the electrostatic potential in the pore is altered by ion binding (Miller, 1999; Nonner et al., 1999; also see Lin and Chen, 2003), it is likely that the internal Cl- controls on the closing rate are also a combination of the above two effects—that is, a direct competition of Cl- with the E166 side chain and the overall more negative potential brought to the pore by Cl- binding. These two mechanisms, however, are probably not separable even though we have intentionally discussed them individually.
Besides an effect on the closing rate of the fast gate, some mutants presented in this study (for example, I515K or D513S) also alter the fast-gate opening rate. Simultaneous increases of the opening and the closing rate by a point mutation could result from a reduction of the transition state energy in a reversible gating process. However, several observations suggest that the fast-gate opening and closing of ClC-0, at the microscopic level, might not be the reverse process of one another. First, the opening and the closing mechanisms are regulated differently by external and internal Cl-. Second, each mutation we have created has a different magnitude effect on the opening and the closing rate. In the present study, in which internal manipulations (such as alterations of [Cl-]i and mutations of residues intracellular to the glutamate gate) are made, changes in the closing rate are always larger than those in the opening rate. On the other hand, extracellular manipulations—for example, alterations of [Cl-]o (Chen and Miller, 1996) and the modification of K165C by MTSEA (Lin and Chen, 2000)—mainly change the opening rate. Finally, the apparent affinity for external Cl- to regulate the opening rate is 50 mM (Chen and Miller, 1996), while the affinity for the internal Cl- to regulate the closing rate is 1.3 M (Fig. 4). These observations together may suggest that protein conformational changes involved in the opening and closing of the fast gate are not the reverse of each other, a possibility consistent with the essence of the nonequilibrium gating cycle proposed previously (Richard and Miller, 1990).
Thus, the functions of the fast gate appear to be more complicated than the very simple form of local motions of an amino acid sidechain as envisioned from the crystal structures of the bacterial ClC channels shown in Dutzler et al. (2003). The competition of Cl- with the negatively charged E166 sidechain may account for the internal Cl- effect on the closing rate, but it is difficult to explain the external Cl- effect on the opening rate. Because the apparent Cl- affinities for external and internal Cl- to control the fast gating are very different, it is unlikely that external Cl- would modulate the opening rate by directly binding to the Sext site, which appears to have an intimate relation with the internal Cl- control of the closing rate. If this is the case, the Cl- binding site for the external Cl--dependent, depolarization-favored, opening of the fast gate should not correspond to any of the three Cl--binding sites revealed in the most recent crystal structure of the E. coli ClC channel (Chen, 2003). It would be interesting to explore the structural basis of the fast-gate opening by extracellular Cl- ions. In these future studies, however, separating the opening from the closing process of the fast gating would be important in order to properly address the irreversible nature of the gating in ClC-0.
|
FOOTNOTES |
|---|
Chia-Wei Lin's present address is Center for Research on Occupational and Environmental Toxicology, Oregon Health and Science University, Portland, OR 97201.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by a Health Science Research Award from UC Davis and a National Institutes of Health grant GM65447.
Olaf S. Andersen served as editor.
Submitted: 7 April 2003
Accepted: 3 October 2003
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Zifarelli, A. R. Murgia, P. Soliani, and M. Pusch Intracellular Proton Regulation of ClC-0 J. Gen. Physiol., July 1, 2008; 132(1): 185 - 198. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.-Y. Chen and T.-C. Hwang CLC-0 and CFTR: Chloride Channels Evolved From Transporters Physiol Rev, April 1, 2008; 88(2): 351 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Engh, J. D. Faraldo-Gomez, and M. Maduke The Role of a Conserved Lysine in Chloride- and Voltage-dependent ClC-0 Fast Gating J. Gen. Physiol., September 24, 2007; 130(4): 351 - 363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X.-D. Zhang, Y. Li, W.-P. Yu, and T.-Y. Chen Roles of K149, G352, and H401 in the Channel Functions of ClC-0: Testing the Predictions from Theoretical Calculations J. Gen. Physiol., March 27, 2006; 127(4): 435 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hinzpeter, J. Lipecka, F. Brouillard, M. Baudoin-Legros, M. Dadlez, A. Edelman, and J. Fritsch Association between Hsp90 and the ClC-2 chloride channel upregulates channel function Am J Physiol Cell Physiol, January 1, 2006; 290(1): C45 - C56. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Engh and M. Maduke Cysteine Accessibility in ClC-0 Supports Conservation of the ClC Intracellular Vestibule J. Gen. Physiol., May 31, 2005; 125(6): 601 - 617. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
, closed level; 
, middle level; and 
, fully open level. The length of the analyzed trace and the number of events in each analysis at closed, middle, and fully open levels, respectively, are: (120 mM) 62 s and 1,147, 1,572, 426; (600 mM) 60 s and 659, 1,251, 593; (2,400 mM) 52 s and 197, 644, 449. (B) Averaged opening rate (left) and closing rate (right) of WT ClC-0 under different [Cl-]i. The rate parameters were calculated according to 
, 1,200; 
, 2,400.
