Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Many physiologically important functions, including regulation of cell volume, transepithelial transport, and electrical excitability rely on members of the large ClC family of voltage-dependent chloride channels (for review, see Jentsch et al., 1999). Mutations in three of these result in human hereditary disorders, including myotonia, Bartter's syndrome, and Dent's disease (Koch et al., 1992; Lloyd et al., 1996; Simon et al., 1997). The skeletal muscle disease myotonia congenita is associated with mutations in CLCN1, the gene encoding the major skeletal muscle chloride channel, ClC-1, and can be inherited in two forms, the autosomal recessive Becker's disease or the autosomal dominant Thomsen's disease (Koch et al., 1992). Almost all mutations resulting in Thomsen's disease exert a dominant negative effect on ClC-1 function and result in a shift in the voltage dependence of ClC-1 gating to more positive potentials (Pusch et al., 1995b; Wollnik et al., 1997; Kubisch et al., 1998). This shift in the voltage dependence of gating accounts for the reduced chloride conductance (G_Cl) seen in myotonic muscle fibers.

Langer and Levy (1968) reported that clofibrate, once commonly used to treat hyperlipidemia, induced a syndrome characterized by muscle stiffness and weakness. In experimental animals, it has been possible to simulate myotonic symptoms, similar to those produced by mutations in ClC-1, using treatment with clofibrate and its analogs (Bryant and Morales-Aguilera, 1971; Dromgoole et al., 1975; Conte Camerino et al., 1988). Two recent studies investigating the effects of 2-(4-chlorophenoxy) propionic acid (CPP), the most potent of the clofibrate analogs, on the wild-type (WT) ClC-1 channel showed that this drug causes changes in ClC-1 properties similar to those found in the myotonic mutant channels: it shifts the voltage dependence of activation of the ClC-1 channel to more positive potentials (Aromataris et al., 1999; Pusch et al., 2000). Because ClC-1 gating depends on Cl binding in the channel pore and P_o of ClC-1 shifts when the external Cl concentration is changed (Rychkov et al., 1996), it was hypothesized that CPP reduces the affinity of the gating site of ClC-1 for Cl (Aromataris et al., 1999). The apparent similarity of action between drug and naturally occurring mutations may indicate that the binding of CPP to its specific binding site produces a change in the channel protein similar to those produced by dominant missense mutations in ClC-1.

The fact that chemical agents can interact with the gating process of ClC-1 suggests that it may be possible to develop a drug that would shift the gating of mutant versions of ClC-1 back toward more negative potentials; therefore, closer scrutiny of the mechanisms of gating are warranted. Recent advances in the area (Saviane et al., 1999; Accardi and Pusch, 2000), and a method for the separation of the P_o of fast and slow gating from the whole-cell currents proposed in the present article has allowed us to investigate further the properties of mutant channels and to compare them with the changes introduced by application of CPP. The results indicate that the F307S and A313T mutations shift P_o of slow gating, as expected from their dominant mode of inheritance, without significant effect on fast gating. CPP, by contrast, mainly affects fast gating in the WT and mutant channels.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Site-Directed Mutagenesis. Point mutations were introduced into hClC-1 cDNA (Steinmeyer et al., 1994) by standard two-step PCR-based site-directed mutagenesis (Ho et al., 1989). PCRs were performed using Pwo DNA polymerase (Roche Molecular Biochemicals, Mannheim, Germany) for high fidelity amplifications. Two fragments were amplified in the first step, using primers containing the desired mutation in a short overlapping region and pTLN-hClC-1 (Lorenz et al., 1996) as a template. In the second step, the two partial overlapping fragments were joined by recombinant PCR. The final products were digested with the appropriate restriction endonucleases and ligated into the pTLN-hClC-1 vector. Restriction endonucleases were used to isolate appropriate fragments carrying the desired mutation in hClC-1, which were then ligated into the pCIneo (Promega, Madison, WI) mammalian expression vector. All PCR-derived fragments were entirely sequenced to exclude any polymerase errors.

Cell Culture and Transfection. Human embryonic kidney (HEK293) cells were grown in Dulbecco's modified Eagle's medium (Invitrogen Australia, Melbourne, Australia) containing 10% (v/v) fetal bovine serum (Trace, Melbourne, Australia), supplemented with L-glutamine (20 mM; Sigma, St. Louis, MO) and maintained at 37°C with 5% CO₂. Twenty-four hours after cell cultures were split, cells were transfected with 0.8 µg of either WT or mutant pCIneo-hClC-1 cDNA using LipofectAMINE PLUS reagent (Invitrogen), following the standard protocol described by the manufacturer, in 25-mm culture wells. To allow ready identification of transfected cells during patch-clamp experiments, cells were cotransfected with ~0.1 µg of green fluorescent protein plasmid cDNA (pEGFP-N1; CLONTECH, Palo Alto, CA). Approximately 6 h after transfection, cells were replated ready for patch clamping. Electrophysiological measurements were commenced 25 h after transfection.

Electrophysiology. Patch-clamp experiments on HEK293 cells were performed in the whole-cell configuration at room temperature (24 ± 1°C) using a List EPC 7 (List, Darmstadt, Germany) patch-clamp amplifier and associated standard equipment. The standard bath solution contained 170 mM NaCl, 2 mM MgCl₂, 2 mM CaCl₂, 10 mM HEPES, adjusted to pH 7.4 with NaOH. The standard pipette solution contained 40 mM CsCl, 110 mM Cs-glutamate 10 mM EGTA-K, 10 mM HEPES, adjusted to pH 7.2 with NaOH. Lower external Cl concentrations were achieved by equimolar substitution of Na-glutamate for NaCl, whereas the high external Cl concentration (i.e., 356 mM Cl) was achieved by doubling the concentrations of all solutes present, except HEPES in the bath solution and HEPES and EGTA-K in the pipette solution. All data presented in the figures were obtained at 178 mM Cl in the external solution, except were specified.

View this table: [in this window] [in a new window]   TABLE 1 Effects of Cl concentration, mutations, and CPP on V1/2 and minimal Po of the fast and slow gates. Both parameters are determined by fitting the Boltzmann distribution to open probabilities calculated from the relative amplitudes of the inward current components.

Chemicals. (R,S)-(±)-CPP was obtained from Sigma. The sodium salt of this compound, which was prepared by neutralizing the corresponding acid with an equimolar amount of NaOH (added as a 1 M solution), was dissolved in freshly made bath solution as required.

Data Analysis. Single channel recording of ClC-1 recently performed by Saviane et al. (1999) is consistent with the presence of two independent gates in this channel: a fast gate that works on each single protopore and a slow gate that gates both protopores simultaneously. Time constants of the fast and slow gates obtained from single-channel recordings are very similar to the time constants of two exponential components that can be fitted to the macroscopic currents. Accepting that the time constants extracted from whole cell currents reflect relaxations of the fast and slow gates, it is possible to derive the open probabilities of fast and slow gates from the relative amplitudes of the corresponding exponential components. During the voltage step from a membrane potential of V₁ to the membrane potential V, the P_o of each gate changes exponentially from one steady state to another. Dependence of the P_o of the fast gate on time can be described by the following equation:

(1)

where P and P are the steady state P_o of the fast gate at the membrane potential V₁ and V, respectively, and _f is the time constant of the fast gate.

(2)

(3)

(4)

t((_s

+ _f)/_s

_f)  e^t/_f

(5)

(6)

(7)

(8)

(9)

(10)

(11)

V

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Kinetics of Inward Current Deactivation and Open Probability of Mutant hClC-1. One of the most characteristic features of the WT ClC-1 channel is that it deactivates with a double exponential time course to a new steady state when stepped to a membrane potential more negative than the Cl equilibrium potential (Fahlke et al., 1996; Rychkov et al., 1996; Accardi and Pusch, 2000). Mutations introduced to different places in the primary structure of the ClC-1 channel can drastically change the kinetics of current deactivation (Fahlke et al., 1997). Substitution of the phenylalanine residue at position 307 with a serine residue or substitution of the alanine at position 313 with a threonine residue resulted in a faster deactivation of the current at negative potentials than in WT ClC-1 (Fig. 1). The 100-ms prepulse to +40 mV was sufficient for maximal activation of WT channel (Fig. 1A), although both mutants required longer prepulses to more positive potentials. Therefore, a different voltage protocol has been used for mutant channels with a 200-ms prepulse to +120 mV followed by voltage steps ranging from 120 to +120 mV in 20 mV increments (Fig. 1, B and C).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Effect of mutations causing dominant myotonia congenita on ClC-1 currents. ClC-1 currents recorded from HEK293 cells expressing WT channel (A), F307S mutant (B), and A313T mutant (C). Voltage protocol: A, prepulse to +40 mV followed by the voltage steps ranging from 120 mV to +80 mV in 20-mV increments; B and C, prepulse to +120 mV followed by the voltage steps ranging from 120 mV to +120 mV in 20-mV increments. Holding potential 30 mV. Currents are normalized to the peak current amplitude at 120 mV.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effect of myotonic mutations on kinetics of inward current deactivation. A, time constants of the fast (1) and slow (2) exponential components of the inward current. Relative amplitudes of the inward current components of the WT and mutant channels: fast exponential component a1 (B); slow exponential component a2 (C); and time-independent component c (D; see eq. 7).

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Effect of myotonic mutations on ClC-1 Po. A, apparent Po curves for WT and mutant channels. Apparent Po was estimated from the peak tail currents after the voltage steps to different membrane potentials (Rychkov et al., 1996; Aromataris et al., 1999). Solid lines represent the Boltzmann distribution (eq. 10). B and C, open probability of the fast and the slow gates, respectively. Data points were calculated from the relative amplitudes of the components of the inward currents as explained under Materials and Methods. In B, solid lines represent the Boltzmann distribution fitted to the calculated data points. In C, solid lines were obtained by dividing the Po curves derived from peak tail currents shown in A by the Po curve of the fast gate shown in B (see eq. 3). For the slow gating of both mutants, data was insufficient for a reliable fit by the Boltzman distribution. Dotted line represents the Boltzmann distribution fitted to the data points calculated for the WT channel.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Dependence of Po of the WT and mutant channels on the external Cl concentration. A, voltage of the half-maximal Po (V1/2) of WT and mutant channels, obtained from tail currents is plotted against the log of the external Cl concentration. The slope of the lines is ~60 mV per decade of Cl concentration change. B, Po curves of the fast and slow gating of the WT ClC-1 in different external Cl concentrations. Solid lines represent the Boltzmann distributions fitted to the calculated data points (eqs. 8 and 9).

Effect of CPP on Fast and Slow Gating in WT and Mutant hClC-1. To determine whether chemically induced myotonia has the same mechanism of action as the naturally occurring myotonia caused by the mutations, we compared the effects of CPP on the WT channel with the effects of the F307S and A313T mutations and investigated the effects of CPP on these mutant channels. Application of 3 mM CPP to the bath solution produced faster inactivating currents from WT channels (Fig. 5A) that were superficially similar to the currents of the mutant channels in the control conditions; and also shifted P_o of the WT channel to more positive potentials by ~50 mV (not shown) as reported previously (Aromataris et al., 1999). Analysis of the relative amplitudes of the inward current components, however, revealed that CPP increased the amplitude of the fast exponential component and decreased the amplitude of the second exponential component (Fig. 5B) in clear contrast to the effects of the mutations on these inward current components (Fig. 2, B and C). The contrasting effects of CPP and mutations on the relative amplitudes of the current components indicate clearly that they have fundamentally different effects on the gating of ClC-1. In fact, application of 3 mM CPP to the WT channel shifted the fast gating toward positive potentials by ~ 60 mV, while shifting slow gating by only ~20 mV (Fig. 5C; Table 1).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Effect of the external application of CPP on the currents and Po of the WT channel. A, wild-type ClC-1 currents recorded in response to a voltage step to 120 mV after a prepulse to +40 mV in control conditions and in the presence of 3 mM CPP in the external solution. Holding potential 30 mV. B, relative amplitudes of the inward current components of the WT ClC-1 inward currents in the control conditions and the presence of 3 mM CPP. (C) Po curves of the fast and slow gating of the WT channel in control conditions and in the presence of 3 mM CPP. Solid lines represent the Boltzmann distributions fitted to the calculated data points (eqs. 8 and 9).

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effect of the external application of CPP on the currents and Po of the mutant channels. A and B, mutant ClC-1 currents recorded in response to a voltage step to 120 mV after a prepulse to +120 mV in control conditions and in the presence of 3 mM CPP in the external solution. Holding potential 30 mV. C, shift of the V1/2 (V1/2) of the WT and mutant channels in the presence of different concentrations of CPP in the external solution. Experimental data points are fitted with a one-site binding hyperbola (eq. 11). D and E, fast and slow gate Po of the mutant channels in the control conditions and in the presence of 3 mM CPP in the external solution. For the fast gate, the Boltzmann distributions were fitted to the calculated data points; for the slow gate, curves were obtained as in Fig. 3C.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The ClC-0 chloride channel from the Torpedo californica electric organ, which is 55% homologous to ClC-1, has been extensively studied on a single channel level and is known for its double-barrelled behavior and two independent gating processes, fast and slow (Hanke and Miller, 1983). Fast gating controls each pore independently on a millisecond time scale, whereas slow gating opens and closes both pores simultaneously on a much longer time scale. Recently, it became apparent from single-channel recording of ClC-1 that it also has two gating processes, similar to ClC-0 (Saviane et al., 1999). These findings showed that the time course of the inward current deactivation reflected the relaxations of the fast and slow gates of ClC-1. Because of its very low conductance, ClC-1 is not easily amenable to single channel recording, and most experiments are restricted to the macroscopic currents. To separate P_o of the fast and slow gating from the macroscopic currents, Accardi and Pusch (2000) used envelope protocols. These envelope protocols could only be used on membrane patches with small capacitance, because they required very short (~200 µs) prepulses to positive potentials. In the present work, we used a different method of separation of the P_o of the fast and slow gating from the whole cell currents. This method and the one based on the envelope protocols give very similar qualitative results (compare Accardi and Pusch, 2000). The biggest difference obtained by these two methods is in the minimal P_o of both gates of the WT channel, which were significantly larger in the study by Accardi and Pusch (2000). The reasons for this could be the different modes employed for recording macroscopic currents: the earlier study used inside-out patches whereas we used the whole cell; in addition, there was a difference in cytoplasmic Cl concentration: 100 mM versus 40 mM. In ClC-0, expressed in Xenopus laevis oocytes, the minimal P_o of the fast gate was significantly lower in the two-electrode voltage clamp than in cell-attached patches and it was also reduced with a lower cytoplasmic Cl concentration (Ludewig et al., 1997).

A dominant mode of inheritance of the mutations causing Thomsen's disease is explained by a dominant negative effect of the mutated subunit on the WT subunit in a multimeric complex (Pusch et al., 1995b; Kubisch et al., 1998). It has been shown that the V_1/2 of the heteromeric mutant/WT complexes is often shifted drastically to more positive potentials (Pusch et al., 1995b) and that the mechanism of this shift lies in the slow gate common to both pores of the ClC-1 channel dimer (Saviane et al., 1999). The results of the present work support this hypothesis; in both mutant channels, F307S and A313T, slow gating was shifted to more positive potentials. Moreover, unlike another dominant mutant, I290M, in which both fast and slow gating were shifted simultaneously (Saviane et al., 1999), mutations F307S and A313T shifted P_o of only the slow gating without much effect on the fast gating mechanism (Fig. 6, D and E). Comparison of the effects of CPP and these dominant myotonic mutations on ClC-1 gating revealed that although they cause similar changes in voltage dependence of ClC-1 P_o, they do not share the same mechanism of action. Unlike F307S and A313T mutations, CPP mainly shifts voltage dependence of the fast gating. These results once more support the notion that the fast and slow gates of ClC channels are different structures and can be manipulated separately. CPP represents an interesting example in this respect. Open probabilities of the F307S and A313T mutants obtained from the normalized tail currents were plainly much less sensitive to CPP than WT. CPP in 3 mM concentration shifted P_o curves in WT by ~50 mV, whereas in F307S, the shift was ~25 mV and in A313T only ~12 mV (Fig. 6C). When P_o of fast and slow gating were separated, it turned out that 3 mM CPP affected fast gating of the mutant channels to the same extent as the fast gating of WT. Therefore, the apparent change of affinity of CPP in the mutant channels could not be caused by a change of the CPP binding to its site. In the mutant channels, the voltage dependence of P_o of different gates are separated to a such an extent that when the slow gate just starts to open, P_o of the fast gate is already close to its maximal value. Consequently, slow gating is the main contributor to the channel's overall P_o obtained from the normalized tail currents, so only the effect of CPP on the slow gate is evident, whereas its effect on the fast gate remains hidden.

A shift in the voltage dependence of channel P_o implies that voltage-dependent steps in channel transition from closed to open states have somehow been altered. A detailed model of ClC-1 gating that would explain all known properties of the channel is yet to be developed, but three models of gating of either ClC-0 or ClC-1 have been proposed. The model for ClC-1 that includes an intrinsic voltage sensor (Fahlke et al., 1996) in the protein structure of the channel is not supported by some of the experimental data, as has been described previously (Saviane et al., 1999; Accardi and Pusch, 2000) and will not be discussed here. Both of the other models proposed to explain voltage and Cl dependence of fast gating of ClC-0 are based on the assumption that the permeating anion also provides the gating charge. Pusch et al. (1995a) suggested that the pore of the channel contains two binding sites for Cl and the fast gate is situated closer to the cytoplasmic side of the channel than the inner site. Occupation of this inner site by Cl shifts the equilibrium between open and closed states to the open state. Voltage dependence of the gating in this model was attributed to the Cl movement from the outer to the inner site. According to this model, alteration of the inner site affinity for Cl or a change in the energy barrier for Cl between two sites could lead to a shift of the P_o along the voltage axes. Change in the inner site affinity for Cl was proposed to be a reason for the shift of ClC-1 P_o in the presence of CPP (Aromataris et al., 1999). However, the recent finding that the time constant of fast gating does not saturate at very high positive potentials, at which Cl binding to the inner site should be saturated (Accardi and Pusch, 2000), is more consistent with a model of the kind proposed by Chen and Miller (1996). The latter model suggested that the binding site for Cl is located externally, and the voltage dependence of channel gating arises from transfer of the bound Cl across the electric field during the conformational change of the channel protein, with the channel opening rate strongly dependent on both voltage and Cl concentration.

Results of the present work taken together with the previous studies could provide some clue about what exactly is changed in mutant channels or in the presence of CPP: the affinity of the gating site for Cl or the free energy of transition between closed and open states. Open probabilities of both gates shift in parallel with changing Cl concentration (Fig. 4B), which suggests that both gates depend on Cl binding to the same site. In this case, alteration of Cl binding to that site will lead to the shift of both fast and slow gating. CPP predominantly shifted fast gating in the WT channel, and the mutations F307S and A313T affected the slow gate, so it is unlikely that the mechanism of their action would be on the Cl binding site; rather, it would be on the voltage dependent transformation to an open state that follows after Cl binding.