Recombinant CLIC1 (NCC27) Assembles in Lipid Bilayers via a pH-dependent Two-state Process to Form Chloride Ion Channels with Identical Characteristics to Those Observed in Chinese Hamster Ovary Cells Expressing CLIC1*

CLIC1 (NCC27) is an unusual, largely intracellular, ion channel that exists in both soluble and membrane-associated forms. The soluble recombinant protein can be expressed in Escherichia coli , a property that has made possible both detailed electrophysiological studies in lipid bilayers and an examination of the mecha-nism of membrane integration. Soluble E. We performed patch clamp electrophysiology on inside-out patches of a CLIC1-expressing CHO cell line using standard methods as reported previously (7). Single-channel recordings from lipid bilayers were obtained using the tip-dip method (11, 12). In brief, patch clamp pipettes (Garner Glass 7052) were made using a P97 Sutter Instruments puller (Novato, CA), coated with Sylgard (Dow Corning, Midland, MI) and fire-polished to a tip diameter of 1–1.5 (cid:1) m and 5–7 megaohm resistance. The same solution was used both in the bath and in the pipette (140 m M KCl, 10 m M Hepes, pH 6). We varied pH by addition of HCl or KOH. As soon as the pipette tip reached the bath solution, a phospholipid monolayer (phosphatidylcholine, Avanti Polar Lipids, Inc., Birmingham, AL) was spread on the surface. The electrode was repeatedly passed through the surface of the solution until the pipette resistance rose above 5 gigaohms. Purified recombinant CLIC1 protein (2 (cid:1) g/ml) was then added to the bath. An Axopatch 1D amplifier and pClamp 7 (both from Axon Instruments, Novato, CA) were used to record and analyze single-channel currents. Current recordings were digitized at 5 kHz and filtered at 800 Hz. coli 30-fold stock into M at either or and left at for 1 h prior to analysis. Far-UV a CD collected over the a Final three and were a 0% Ficoll buffer-only top layer that filled the remainder of the centrifuge tube. The gradient was centrifuged for 2 h at20,000 rpm in an SW41 swing-bucket rotor at 5 °C. The unincorporated protein was collected from the 40% Ficoll fraction remaining at the bottom of the gradient, and the liposomes were collected from the 30% Ficoll/0% Ficoll interface. The collected liposome fractions were concentrated to a 50- (cid:1) l volume using a RediVac, being careful not to dry the samples completely, and resuspended in 350 (cid:1) l of SDS (3 g of SDS/100 ml of H 2 O). 30 (cid:1) l of each sample was mixed with 10 (cid:1) l of reducing 6 (cid:5) Laemmli sample buffer with 20% (cid:2) -mercaptoethanol, boiled for 5 min, and immediately electrophoresed on a 10% polyacryl- amide gel. CLIC1 was detected by Western blotting as described previously (1).

CLIC1 (formerly NCC27) is a 241-amino acid ion channel protein, which we first identified by subtraction cloning, because of its increased expression with macrophage activation. CLIC1 was the first identified human member of a growing family of organellar ion channels (1). At least five other members of this family have now been cloned, and although they share significant sequence similarity (about 60 -75%), they vary in their cellular and subcellular distribution (2)(3)(4)(5)(6). CLIC1 is widely distributed, being present in most tissues and cells that have been thus far examined. Immunofluorescence studies of CLIC1-transfected CHO 1 cells indicate that it dominantly localizes to the nucleoplasm and nuclear membrane as well as the cytoplasm (1). However, it has also been visualized within small intracytoplasmic organelles, presumably lysosomes, endosomes, or secretory vesicles (1).
CLIC1 can be identified electrophysiologically on the nuclear and plasma membranes in CLIC1-transfected cells and exhibits similar characteristics in both locations (1,7,8). In normal untransfected CHO cells, this translocation of CLIC1 to the plasma membrane occurs in a cell cycle-dependent manner (8). Native CLIC1 channel activity is routinely found only on the plasma membrane of cells in the act of dividing or cells that have just divided (7,8). However, when cells are transfected with CLIC1, this normal distribution pattern is disrupted, and CLIC1 Cl Ϫ conductance can be detected on the surface of all cells (7,8). This suggested a role for CLIC1 in the cell cycle. This view is further reinforced by the observations that CLIC1 blockers increase the percentage of CHO cells in the G 2 /M phase of the cell cycle (8).
CLIC1 is quite unusual for an ion channel: it is small with only a single putative transmembrane region; it is largely a soluble intracellular protein with only small amounts of protein being found on membranes; and the structure of its soluble form does not resemble that of an ion channel. We have recently determined the high resolution, x-ray crystallographic structure of the soluble form of CLIC1 (9). The protein is * This work was supported by grants from the National Health and Medical Research Council of Australia; St. Vincent's Hospital, Sydney; and Meriton Apartments Proprietary Ltd. through an R&D Syndicate arranged by Macquarie Bank Limited and by a New South Wales Health Research and Development Infrastructure grant. This work has been also supported by Italian Ministry of University and Research (MURST) and by intramural funds of "La Sapienza" University. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
b These authors contributed equally to this work. monomeric, structurally homologous to the GST superfamily and has a redox-active site resembling glutaredoxin. Integration of CLIC1 into a membrane is likely to require a major structural rearrangement, probably of the N-domain.
Despite its unusual properties, we and others have previously provided strong data identifying it as an ion channel (1,7,8,10,11). To further investigate the electrophysiological characteristics of CLIC1 and to better understand its regulation, we describe a series of biochemical, biophysical, and electrophysiological studies performed using soluble, Escherichia coli-derived, recombinant CLIC1. We demonstrate that the isolated protein is able to form functional ion channels in an artificial lipid bilayer and that these channels have essentially identical electrophysiological characteristics to those seen in cell membranes. Channel formation involves at least two stages: the initial formation of a small conductance slow kinetics (SCSK) channel followed by the appearance of a high conductance fast kinetics (HCFK) channel, where the latter corresponds to the in vivo channel. Our results indicate that the HCFK channel probably results from the assembly of four SCSK conductances. Finally, we demonstrate that the association of CLIC1 with lipids and ion channel formation in bilayers are pH-dependent.

EXPERIMENTAL PROCEDURES
Expression of Recombinant CLIC1-A recombinant GST-CLIC1 fusion protein was expressed in E. coli using the pGEX-4T-1 vector system and purified as described previously (1,9). Briefly, the fusion protein was immobilized on glutathione S-Sepharose (Amersham Biosciences), where it was cleaved with biotin-labeled thrombin (Novagen). The thrombin was removed with a streptavidin-agarose affinity matrix, and CLIC1 was further purified by gel filtration on a Superdex G75 column (Amersham Biosciences), where it ran as a monomer.
Electrophysiology-We performed patch clamp electrophysiology on inside-out patches of a CLIC1-expressing CHO cell line using standard methods as reported previously (7). Single-channel recordings from lipid bilayers were obtained using the tip-dip method (11,12). In brief, patch clamp pipettes (Garner Glass 7052) were made using a P97 Sutter Instruments puller (Novato, CA), coated with Sylgard (Dow Corning, Midland, MI) and fire-polished to a tip diameter of 1-1.5 m and 5-7 megaohm resistance. The same solution was used both in the bath and in the pipette (140 mM KCl, 10 mM Hepes, pH 6). We varied pH by addition of HCl or KOH. As soon as the pipette tip reached the bath solution, a phospholipid monolayer (phosphatidylcholine, Avanti Polar Lipids, Inc., Birmingham, AL) was spread on the surface. The electrode was repeatedly passed through the surface of the solution until the pipette resistance rose above 5 gigaohms. Purified recombinant CLIC1 protein (2 g/ml) was then added to the bath. An Axopatch 1D amplifier and pClamp 7 (both from Axon Instruments, Novato, CA) were used to record and analyze single-channel currents. Current recordings were digitized at 5 kHz and filtered at 800 Hz.
Circular Dichroism and Fluorescence Spectroscopy-Purified, E. coliderived recombinant CLIC1 was diluted 30-fold from stock solutions of about 5 mg/ml into 20 mM potassium phosphate buffer at either pH 7.0 or pH 6.0 and left at room temperature for 1 h prior to analysis. Far-UV CD spectra were recorded on a Jasco J-720 spectropolarimeter equipped with a Neslab RTE-111 temperature controller. CD data were collected over the wavelength range of 195-260 nm and with a resolution of 0.5 nm, a bandwidth of 1 nm, and a response time of 1 s. Final spectra were the sum of three scans accumulated at a speed of 20 nm⅐min Ϫ1 and were baseline-corrected.
Fluorescence data were collected on a PerkinElmer Life Sciences LS50B luminescence spectrophotometer using 0.5-cm path length cells and 10-nm slit widths. The temperature was maintained at 25°C using a Bio-Rad E4850 refrigerated recirculator water bath. The excitation wavelength was 295 nm (to specifically excite tryptophan fluorescence), and the emission was monitored from 320 to 380 nm.
Liposome Preparation-Liposome synthesis was based on a method from Hase et al. (13) modified as follows: phosphatidylcholine (Sigma) and cholesterol (Sigma) were dissolved in chloroform (Ajax Chemicals) at 100 and 10 mg/ml, respectively, and then combined to give a final 9:1 phosphatidylcholine:cholesterol ratio (18 mg of phosphatidylcholine ϩ 2 mg of cholesterol). The lipid mixture was then coated in layers onto the sides of a glass test tube by rotating the tube under a stream of nitrogen. The lipids were dried overnight under vacuum and then resuspended by vortexing in 1 ml of HK buffer (140 mM KCl, 10 mM HEPES, pH adjusted with HCl). The lipid suspension was sonicated with glass beads in a Branson 3200 water bath sonicator until it became translucent (ϳ5-10 min), which is indicative of small, unilamellar liposomes having been formed.
Measurement of Association of CLIC1 with Liposomes-To measure association of CLIC1 with liposomes, 1.8 g of purified recombinant CLIC1 was incubated with 150 l of liposomes for periods between 10 min and 4 h. In experiments to check the effect of decreased pH, CLIC1 was combined with liposomes, and the pH was then adjusted with HCl. Liposomes were then purified on a discontinuous Ficoll density gradient as follows: liposomes were adjusted to 40% Ficoll and 1 ml of final volume with HK buffer and layered at the bottom of the gradient beneath 3 ml of 30% Ficoll and a 0% Ficoll buffer-only top layer that filled the remainder of the centrifuge tube. The gradient was centrifuged for 2 h at 20,000 rpm in an SW41 swing-bucket rotor at 5°C. The unincorporated protein was collected from the 40% Ficoll fraction remaining at the bottom of the gradient, and the liposomes were collected from the 30% Ficoll/0% Ficoll interface. The collected liposome fractions were concentrated to a 50-l volume using a RediVac, being careful not to dry the samples completely, and resuspended in 350 l of SDS (3 g of SDS/100 ml of H 2 O). 30 l of each sample was mixed with 10 l of reducing 6ϫ Laemmli sample buffer with 20% ␤-mercaptoethanol, boiled for 5 min, and immediately electrophoresed on a 10% polyacrylamide gel. CLIC1 was detected by Western blotting as described previously (1).

Soluble Recombinant CLIC1 Protein Forms Cl Ϫ Channels in Lipid Bilayers with Essentially Identical Biophysical
Characteristics to Clic1 on CHO Cell Membranes CLIC1 Forms Cl Ϫ Channels in Artificial Bilayers-Despite its small size and aqueous solubility, CLIC1 can be readily observed to form Cl Ϫ channels in tip-dip bilayer experiments. Fig. 1 shows two examples of single-channel recordings obtained at two different pipette potentials (12,14). Comparable results were obtained regardless of the potential used. Channel recordings always showed the same sequence of events. After bilayer formation on the electrode tip and addition of protein to the external solution, there was a variable time period of null events. Channel activity then started with a few long openings that, during the progression of the experiment, became more consistent in number. In 80% of successful experiments, multiple events turned into one or two larger size channels (as seen in Fig. 1, on the right of the traces) in which not only does the amplitude of each individual current step increase, but also the open and close kinetics turn into fast transitions (flickering).
The analysis of such current recordings is very complicated because even superficial observations indicate that we are dealing with two different ionic conductances. The initial portion of a typical current trace shows the predominance of a small In the example, we show single-channel current recordings at 50-and 60-mV pipette potential.
conductance channel with slow kinetics, which we call SCSK. These SCSK events show multiple current levels or steps. During a typical experiment, these SCSK events disappear, and they are replaced by a high conductance channel with fast kinetics, which we call HCFK. We have analyzed the SCSK pathway separately from the HCFK ionic pathway. Fig. 2 shows a detailed analysis of the SCSK channel seen in tip-dip experiments. Both inward and outward current recordings obtained at Ϯ80 mV in the patch pipette are plotted on the left of the figure. In the center column, we show current amplitude histograms collected over three consecutive seconds of recording. The i/V relationships depicted on the right were plotted using the histogram peak with the highest current from each experiment. The conductances obtained by linear regression were 7.7 Ϯ 0.95, 13 Openings of SCSK channel are prominent in tip-dip experiments but can also be found in current recordings from the plasma membranes of CHO cells that have been transfected to express CLIC1 (called CHO-CLIC1). In Fig. 3, we show several examples of single-channel recordings in equimolar 140 mM KCl obtained in tip-dip configuration (left) and in inside-out patches from CHO-CLIC1 (right). In inside-out current traces, smaller current levels are only occasionally evident as part of the single-channel kinetics. Observing native channels, the SCSK type of openings (as judged by their amplitude) show both fast and slow kinetics. This is probably due to the fact that channel subunits are assembled to form a HCFK channel, and current substates are sporadic events occurring randomly. In tip-dip experiments, SCSK openings are produced by proteins that have not yet assembled into a higher order HCFK channel. The recordings presented on the right side of Fig. 3 were chosen from a 22-min continuous recording. By contrast, tip-dip channel current traces clearly showed many small current transitions and (rarely) simultaneous multiple openings or closings in the initial portion of an experiment. However, as soon as the activity of the HCFK channel appears in the tip-dip experiment, SCSK current openings became sporadic, and their frequency of appearance was then comparable with those in inside-out patch experiments.
We note that the conductance of the SCSK channel is 30.5 pS when four current levels are present. This is almost identical to the conductance of the HCFK channel reported previously (9) in both the tip-dip experiments (31.2 Ϯ 1.5 pS) and the recordings of CLIC1 from CHO-transfected cells (29.6 Ϯ 1.9 pS).
In light of these results, we have reason to believe that in the tip-dip experiments, the activity of HCFK channel represents the main conductive state of the fully reconstituted CLIC1 channel, and the SCSK pathway is due to substates of the same channel visible only during the channel assembly process. To explore the main channel characteristics, we will then consider only the openings of HCFK channels. Accordingly, Fig. 4 shows a series of shorter single-channel current recordings obtained in equimolar 140 mM KCl at different patch-pipette potentials. They produced the current/voltages relationship shown on the right of the Fig. 4. The mean single-channel conductance for the HCFK was 31 Ϯ 1.8 pS (n ϭ 5).

CLIC1 in Bilayers Forms an Ion Channel with Similar Kinetic Characteristics to Those in CHO Cell
Membrane-It is critical to our understanding of CLIC1 to determine whether its electrophysiological characteristics in lipid bilayers are similar or identical to those in cell membranes. Analysis of the CLIC1 HCFK pathway in lipid bilayers indicates a conductance value of 30.5 pS. In our recent experiments (9), both tip-dip HCFK pathway and inside-out recordings of CLIC1 from CHOtransfected cells (CHO-CLIC1) showed conductances of 31.2 Ϯ 1.5 pS and 29.6 Ϯ 1.9 pS, respectively. This indicates that CLIC1 has essentially identical conductance characteristics on the cell membrane as in lipid bilayers.
To further compare the characteristics of this ion channel in tip-dip bilayer and in CHO-CLIC1 membranes, we analyzed the kinetic characteristics of HCFK single-channel recordings in the two preparations. Fig. 5 depicts open probability (P o ) and open-and closed-time histograms (the example is at V p ϭ Ϫ40 mV) for HCFK single channels in both experimental configurations. P o was very similar in the two models (0.478 Ϯ 0.035 for inside-out and 0.532 Ϯ 0.026 for tip-dip) and remained constant over a wide range of membrane potentials (Fig. 5, left). In the experiments shown in Fig. 6a, we applied a Ϯ100-mV voltage ramp to artificial bilayers in which only one HCFK CLIC1 channel was active. Using this voltage ramp protocol, a slight rectification of the channel at high potentials is sometimes evident (11). Using the same voltage protocol, we com-pared the Cl Ϫ dependence of CLIC1 conductance in the insideout configuration obtained from CHO-CLIC1 (b), which we have reported previously (7), with that produced using the recombinant protein in lipid bilayers (c). The channel conductance in artificial membranes drops to 13 Ϯ 2.3 pS (n ϭ 5) in 20 mM external Cl Ϫ with a reversal potential of Ϫ60 mV. This is in good agreement with that which we have reported for CHO-CLIC1 (7). Also clearly evident is the shift in current reversal potential in Fig. 6, b and c, as compared with Fig. 6a. Channel openings are seen in Fig. 6, b and c, at the 0 mV holding potential before and after the ramp. By contrast, in Fig. 6a, in equimolar Cl Ϫ conditions, no ionic current is flowing during these recording segments. The outward channel openings in Fig. 6, b and c, demonstrate a negative shift of the current reversal potential.
CLIC1 Current Produced in Artificial Bilayers Is Blocked by IAA94 -The CLIC1 ion channel, when expressed in CHO cells, is inhibited by IAA94 and A9C but not DIDS (7). We wanted to determine whether CLIC1 in artificial bilayers exhibited the same characteristics. We found that CLIC1 channels in bilayers have an identical inhibitor profile. Since the blocking action of A9C is irreversible, we confined subsequent studies to the use of IAA94. Fig. 7 (top) shows the effect of increasing IAA94 concentration during perfusion of a tip-dip patch containing only one functional HCFK Cl Ϫ channel clamped at Ϫ60 mV. Increasing the concentration from 1 to 50 M results in a dose-dependent decrease in the channel open probability. In the dose/response graph (Fig. 7, bottom), the normalized channel open probability, P o , is plotted against the IAA94 concentration. This indicates a maximum effect at concentrations between 50 and 100 M with a Hill coefficient of 1.4 and an IC 50 of 25 M (n ϭ 4). All these data are in good agreement with our own data from outside-out experiments performed on CHO cells (8) and with the bilayer experiments reported by Tulk et al. (11).
In a number of experiments (for example, Fig. 8), many SCSK current sublevels were observed without the formation of the HCFK channel. We found that blockade of these higher order SCSK substates required concentrations of IAA94 from 5 to 10 times higher than those for the unitary HCFK conductance (above 300 M).

Effect of pH on Both Electrophysiological and Physical
Characteristics of CLIC1 Decreasing pH Facilitates the Appearance of CLIC1 Channels in Artificial Membranes-To determine whether ion channel formation in artificial membranes was pH-dependent, we added recombinant CLIC1 to the bath solution at various pH levels and then monitored the time to detect the first event. In Fig. 9, we show three examples of this procedure with the pipette potential held at Ϫ50 mV. In all three cases, the initial ion channel activity appears as SCSK channels, as described previously in the legend for Fig. 1. However, as the pH decreases, there is a decrement of the time to observe the first opening. In addition, the probability of observing a 30-pS channel (HCFK) was higher at low pH. In Fig. 10, we have plotted both the average time before the appearance of the first conductance (A), usually an SCSK channel, and the probability of forming a complete HCFK channel (B). The plot on the left of Fig. 10 clearly suggests the existence of a threshold between pH 6 and 6.5 in which the channel changes drastically its ability to insert into the artificial membrane. The time needed to observe a functional SCSK ion channel is almost tripled as the pH of the bath solution increased from pH range 5-6 to range 6.5-7. Probably as a consequence, the histogram plot on the right shows that the formation of HCFK channels has a higher probability at low pH. However, all attempts to determine whether pH is also involved in the transition of single functional SCSK units to form a HCFK channel in the membrane have failed. A simple explanation of our observations could be that low pH increases the probability of protein insertion into the lipid membrane, increasing the concentration of CLIC1 in the bilayer. This, in turn, favors the aggregation of channel SCSK subunits into a HCFK channel. We have no data at the moment to show a direct influence of the change in the pH of the solution on multimer formation in the membrane.
Soluble, Recombinant CLIC1 Interacts with Artificial Liposomes-To observe the interaction of CLIC1 with artificial lipid bilayers, purified recombinant CLIC1 was incubated with liposomes for between 10 min and 4 h. The liposomes were then separated from any remaining soluble protein by purification on a discontinuous Ficoll density gradient. The liposome-con-taining fraction was analyzed for CLIC1 content by Western blotting, as was the dense bottom fraction corresponding to the part of the gradient where the sample was loaded (Fig. 11A). To verify that soluble CLIC1 did not migrate alongside the liposomes, CLIC1 was also centrifuged through the gradient in the absence of liposomes, and the fraction where the liposomes would normally be found was analyzed by Western blotting and found to be free of the protein (data not shown).
The liposome fraction (lane 1) contains a substantial amount of CLIC1, indicating that the soluble protein does associate with artificial lipid bilayers as determined biochemically. The fraction in which the sample was loaded on the gradient (lane 2) contains CLIC1, which had not associated with the liposomes and remained in a soluble state.
Low pH Facilitates the Interaction of CLIC1 with Liposomes-Measurement of the association of CLIC1 with liposomes at pH 6.5 and pH 7.4 indicates that low pH facilitates the interaction of CLIC1 with lipid bilayers (Fig. 11B). To determine the effect of low pH on CLIC1 association with lipid bilayers, CLIC1 was combined with liposomes, and the pH of the mixture was adjusted to pH 7.4 or pH 6.5. The mixture was incubated between 10 min and 4 h, and the liposomes were purified by centrifugation on a discontinuous Ficoll density gradient, which also had the pH adjusted accordingly. Following centrifugation, the liposome fraction was collected, and the amount of CLIC1 it contained was determined by SDS-PAGE and Western blotting followed by densitometry. CLIC1-lipo-some association was found to increase in a time-dependent manner and to be enhanced at pH 6.5 as compared with pH 7.4 (Fig. 11B).
Low pH Does Not Cause a Major Alteration in the Secondary Structure of Soluble CLIC1-To determine whether the pH-dependent effects on CLIC1 ion channel activity were associated with a conformation change in the protein, we first looked for any changes to secondary structure using CD spectropolarimetry. The spectrum obtained for a CLIC1 in potassium phosphate buffer at pH 7.0 is typical of a protein containing a significant proportion of ␣-helical structure with minima at 208 and 222 nm (Fig. 12). No differences were observed when the protein was incubated at pH 6.0, suggesting that there are no significant changes in the proportions of secondary structural elements of CLIC1 in this pH range.
To determine whether there were any structural changes that might involve a rearrangement of secondary structural elements, rather than a change in total secondary structure, we looked for changes in tryptophan fluorescence between pH 7.0 and pH 6.0. Changes in the emission wavelength maximum ( max ) or the intensity of fluorescence would indicate a change of environment of the single tryptophan residue at position 35 in CLIC1. At both conditions, CLIC1 exhibited a max of 345.5 nm, which is typical for a partially buried tryptophan residue. This suggests that there are no pH-induced structural changes associated with Trp-35. DISCUSSION In the present study, we were able to characterize not only the biophysical characteristics of CLIC1 protein, as the essential component of a Cl Ϫ -selective anion channel, but also the sequence of events relating to the formation of a functional single channel comparable with the native conductance. The temporal sequence, following addition of protein to the trans solution, was always similar and is well represented in Fig. 1. After an initial delay, SCSK conductances appear in the bilayer. These SCSK modules then appear to undergo a transition to form the HCFK channel, which has four times the conductance of the SCSK and fast kinetics that correspond to the characteristics of the native channel.
Our current model is based on the assumption that either a single 241-amino acid CLIC1 protein or a well defined multimer of CLIC1 is able to span the membrane and to open a channel with very slow opening and closing kinetics, the SCSK conductance. Thus the current traces in most of our experiments are characterized, in particular at the beginning of recording, by the presence of many current levels. Eventually single proteins (or well defined multimer SCSK units) are able to assemble in a complex that duplicates the fast open and closed transition typical of the native CLIC1 (1,7,8). According to our data and calculations and based on the conductance measurements (Fig. 6), such a channel needs four SCSK ionic pathways combined to equate in amplitude to a wild-type CLIC1 channel.
This process is not a simple aggregation of subunits; the suggested four elements must also cooperate in determining the opening and closing kinetics of the resultant Cl Ϫ channel. This complementary effect of the single subunits is also manifest during the action of the specific channel blocker IAA94. A channel (HCFK) similar to the native form displays an IC 50 of 25 M (Fig. 7). We were not able to calculate the same parameter for the SCSK channel; from our results, it appears that with a drug concentration up to 100 M, openings remain unchanged (Fig. 8)   probability of the channels (n ϭ 6). More importantly, although the 30-pS channel shows prompt reversibility of the IAA94 current block upon washout, we were never able to observe any more openings of the small SCSK channels after they were closed by addition of IAA94. This could be interpreted in several ways.
FIG. 10. CLIC1 single-channel activity versus pH. As shown in A, the time of the first functional insertion in the lipid bilayer increases markedly as the solution pH goes from a range of 5-5.5 to one of 6.5-7. B, probability of CLIC1 protein forming a 30-pS high current fast kinetics HCFK channel. The histogram shows a higher occurrence of channel aggregates at lower pH. However, it is impossible to distinguish whether the pH is responsible for this phenomenon or whether the higher number of sublevels observed at low pH contributes by itself to increase the probability of 30-pS channel formation.
FIG. 11. CLIC1 associates with artifical liposomes. A, Western blot of fractions collected following discontinuous Ficoll density gradient purification of liposomes: a purified liposome fraction was collected from the 30% Ficoll-0% Ficoll interface, which contains liposome-associated CLIC1 (lane 1). Soluble CLIC1 that did not migrate through the gradient was collected from the bottom fraction where the protein/liposome mixture was loaded (lane 2). CLIC1 associates with liposomes, and relative to the soluble fraction, the liposome fraction contains more multimers. B, time course of CLIC1 association with liposomes. This was determined over 4 h, at pH 7.4 and pH 6.5, by Western blotting of Ficoll gradient-purified liposome fractions followed by densitometry. At pH 6.5, more CLIC1 is present in the liposome fractions than at pH 7.4, indicating that low pH facilitates CLIC1-lipid bilayer interaction. Inset, Western blot of time course of CLIC1 association with liposomes at pH 7.4 and pH 6.5; time points are between 10 min and 4 h, as indicated on graph.
FIG. 12. Circular dichroism of CLIC1. CLIC1 was diluted from a stock solution into potassium phosphate buffer at either pH 7.0 (solid line) or pH 6.0 (dotted line) and incubated for 1 h at room temperature, and then CD data were collected over the wavelength range 195-260 nm. The spectra shown were the sum of three scans and were baseline-corrected.