| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Department of Neurobiology, Pharmacology, and Physiology, The University of Chicago, Chicago, IL 60637, USA2 Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH 45267, USA3 Department of Pediatrics, University of Iowa, Iowa City, IA 52242, Chicago, IL 60637, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 12 November 2003;
accepted after revision 28 January 2004;
first published online 30 January 2004)
Corresponding author D. J. Nelson: Department of Neurobiology, Pharmacology, and Physiology, The University of Chicago, 947 East 58th Street, AB-500 MC-0926, Chicago, IL 60637, USA. Email: dnelson{at}drugs.bsd.uchicago.edu
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
While broad expression and physiological importance of CLC-3 are established, the mechanism of channel activation remains controversial. CLC-3 has been proposed to be the ubiquitous volume-regulated channel (Duan et al. 1997; Yamazaki et al. 1998; von Weikersthal et al. 1999). However, experiments in cells lacking CLC-3 have concluded that CLC-3 is not the swelling-activated chloride channel (Li et al. 2000; Huang et al. 2001; Stobrawa et al. 2001; Weylandt et al. 2001). Additionally, it has been reported that CLC-3 is Ca2+ regulated, either via a direct pathway (Kawasaki et al. 1995) or through the multifunctional Ca2+–calmodulin-dependent protein kinase II (CaMKII) (Huang et al. 2001), while a knockout study reports no loss of an ionophore-induced Ca2+-activated Cl– conductance (Arreola et al. 2002). It should be noted that CLC-3 has multiple splice variants and the differential expression of each of these in various cell types may lead to diverse regulation of the channel.
Regulated Ca2+-dependent Cl– conductances have been described in a diversity of cell types, including neurones, smooth muscle and secretory epithelia. The multifunctional CaMKII is a major mediator of Ca2+ signalling and is abundant in the brain, accounting for approximately 2% of total protein in the hippocampus (Erondu & Kennedy, 1985). Phosphorylation-dependent gating of Cl– channels by CaMKII is well established in many cell types (Wagner et al. 1991; Worrell & Frizzell, 1991; Chan et al. 1994; Holevinsky et al. 1994; Kaetzel et al. 1994; Xie et al. 1996, 1998). Previous work in our laboratory has demonstrated that CLC-3 exhibits CaMKII-dependent channel gating in a stable cell line expressing recombinant human CLC-3 and in T84 cells expressing CLC-3 endogenously. In both cell types, the currents activated by autonomously active CaMKII are identical in their biophysical properties. While CaMKII activates the long form of human CLC-3 (hCLC-3) when expressed at the plasma membrane, the mechanism of channel activation when expressed in the cytoplasmic compartment remains uninvestigated.
High levels of CLC-3 expression in the brain, kidney and colon emphasize the cellular and physiological importance of this channel in secretory processes. In the present study, we set out to establish the molecular identity of the endogenous CaMKII-activated Cl– current (ICl,CaMKII). Using a combination of both molecular biological as well as electrophysiological studies, we determined that ICl,CaMKII native in smooth muscle and transfected tsA cells is due to the activation of CLC-3. The channel localizes to both cytoplasmic as well as a plasma membrane site in HT29 cells, where its activation by the kinase is not dependent upon a translocation event between the two compartments. Furthermore, the serine at position 109 in hCLC-3 is necessary for phosphorylation and kinase activation of CLC-3, and mutation of this serine to an alanine results in a loss of CaMKII-activated Cl– conductance. Taken together, these data indicate that CLC-3 expressed at the plasma membrane is a Cl– conductance activated by CaMKII-dependent phosphorylation at S109.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Primary tissue explants were made from young adult CLC-3–/– and CLC-3+/+ mouse aortae using established methods (Freshney, 2000). As approved through the University of Iowa Animal Care and Use Committee (Protocol no. 0301002 ‘CLC-3 chloride ion channels in vascular smooth muscle’), the CLC-3–/– and CLC-3+/+ mice were killed by exposure to high CO2 followed by cervical dislocation, without the use of anaesthetic. Expression of smooth muscle 
-actin was demonstrated in each cell line by immunostaining (antismooth muscle -actin, Sigma, MO, USA; data not shown). Both aortic smooth muscle cells and a human colonic tumour cell line, HT29 (American Type Culture Collection, VA, USA), were grown in Dulbecco's modified Eagle's medium (Invitrogen, CA, USA); tsA cells stably transfected with CLC-3 were grown in minimum essential medium (Invitrogen, CA, USA). Fetal bovine serum (10%, Hyclone, UT, USA) and penicillin–streptomycin (1%, Sigma) were added to the medium, and cells were maintained at 37°C in 5% CO2.
For electrophysiological studies, cells were used 1–2 days after plating or transfection of 35 mm dishes (Corning, Fisher, IL, USA). For immunofluorescence studies, cells were plated on 25 mm glass coverslips (Fisher) treated with poly-D-lysine (Sigma), and used 2 days following plating or transfection. For immunoprecipitation studies, cells were used 2 days after plating or transfection on 100 mm dishes (Fisher).
Cells were transiently transfected (Lipofectamine, Invitrogen) with 1 or 2 µg DNA construct (Flag-CLC-3, Flag-S109A-CLC-3) or no DNA (mock-transfected) per 25 mm coverslip or 35 mm plate. Cells were incubated for 2 days before experiments. Control cells were not transfected.
Immunoprecipitation of CLC-3
Following the protocol provided by Upstate Biotechnology (NY, USA), cells were cultured on 100 mm dishes, lysed in 1 ml of modified radio immuno precipitaton (RIPA) lysis buffer (mM: Tris-HCl 50; pH 7.4, NaCl 150, nonidet P-40(NP-40) 1%) containing proteinase inhibitors. One-third of the volume of lysate was diluted with phosphate-buffered saline (PBS) to 1 ml, precipitated with 20–50 µg of -hCLC-3730–744, then collected with 100 µl of 50% recombinant protein A agarose (Sigma). Immunoprecipitated protein was incubated at 37°C for 15 min in 1% SDS sample buffer containing 5 mM dithiothreitol. The membrane was immunoblotted with -hCLC-359–74 at a 1 : 500 dilution. The CLC-3 antibodies used were those previously described (Huang et al. 2001).
Protein obtained from homogenates of aortae or cultured aortic vascular smooth muscle(VSM) cells were used to identify native CLC-3 protein as previously described (Dickerson et al. 2002). Blots were incubated overnight at 4°C in a 1 : 1000 dilution of rabbit -CLC-3 polyclonal antibody (Alomone Laboratories, Jerusalem, Israel), washed 5 times and incubated for 1 h at room temperature in a 1 : 30 000 dilution of horseradish peroxidase-conjugated goat -rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), and again washed 5 times. Immune complexes were detected on film using enhanced chemiluminescence (Amersham Pharmacia Biotech, Pisscataway, NJ, USA).
In vitro phosphorylation
Whole cell.
tsA cells stably transfected with CLC-3 were solubilized in modified RIPA buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP-40) with proteinase inhibitors. Protein A-conjugated sepharose (100 µl) was saturated with -hCLC-3730–744 (1.5 mg) and used to immunoprecipitate CLC-3 from the cell lysate at 4°C overnight. Following two washes with PBS and one wash with buffer (20 mM Mops pH 7.2, 25 mMß-glycerol phosphate, 1 mM MnCl2, 1 mM CaCl2, 1 mM Na3VO4, 1 mM NaF, 1 mM DTT, and proteinase inhibitors), the precipitated CLC-3 was incubated with 50 µl of CaMKII reaction mixture (CaMKII 2 µg, CaCl2 5 mM, CaM 3 µg, [
-32P]ATP 10 µCi, Mg-ATP 7 µl) at 30°C for 30 min. Rabbit synapsin (2 µg) was used as a positive control. The phosphorylation reaction was terminated and CLC-3 was eluted with Western blot sample buffer. The supernatant was resolved on a 7.5% SDS-PAGE. The resulting bands on the gel were detected using autoradiography.
GST fusion proteins. The cDNA encoding the N- (1M to 122L) or C-terminus (661R to 818 N) of human long form CLC-3 was subcloned into pGEX-6p-1 (Pharmacia) by the restriction digestion with BamH I/EcoR I, and EcoR I/Xho I, respectively. The constructs were transformed into Escherichia coli strain BL21. To express the N-terminal Glutathione S-transferase (GST) fusion protein, bacteria were grown at 29°C in 2 yeast tryptine (2-YTA) medium to an optical density of 0.4 at OD600, then induced with 0.1 mM of isopropyl-beta-D-thiogalactopyranoside (IPTG) for 19 h. For the C-terminal fusion protein, bacteria were grown at 29°C in 2-YTA medium to OD600 reaching an optical density of 0.9 before induction with 0.03 mM IPTG for 2 h. Cells obtained from 1 l cultures were harvested and incubated in 100 ml lysis buffer (PBS, pH 8.0, 0.1 mg ml–1 lysozyme, 10 mM DTT, 1 mM phenylmethylsulphonyl fluoride (PMSF), 1 mM EDTA, 0.1 mM benzamidine, 1 µg ml–1 each (pepstatin A, apotinin, leupeptin) for 30 min on ice prior to sonication. The lysates were clarified twice by centrifugation at 40 000 g for 15 min at 4°C. GST fusion proteins were purified with glutathione–sepharose 4B beads following the batch purification protocol provided by Pharmacia, eluted with 10 mM reduced glutathione in 50 mM Tris buffer, pH 8.5, concentrated with Centricon YM-10 (Fisher Scientific), and dialysed with Slide-A-Lyser Mini Dialysis Units, 7000 MWCO (Pierce) against PBS, pH 7.4. Proteinase inhibitors were added and the product was stored at –80°C.
For in vitro phosphorylation, the N- or C-GST-CLC-3 fusion proteins (15 µg) were incubated with 50 µl of CaMKII reaction mixture (CaMKII 2 µg, CaCl2 5 mM, CaM 3 µg (-32P) ATP 10 µCi, Mg-ATP 7 (µl) at 37°C for 5 min. Rabbit synapsin (2 µg) was used as a positive control. The phosphorylation reaction was terminated and CLC-3 was eluted with Western blot sample buffer. The supernatant was resolved on a 12.5% SDS-PAGE. The resulting bands on the gel were detected using autoradiography.
Construction of Flag-CLC-3
The Flag epitope (DYKDDDDK) was inserted in the hCLC-3 pcDNA3.1 expression vector via PCR utilizing primers (5'-GTCGTCGTCGTCTTTGTA-3' and 5'-CAGCAGCAGCAGAAACATCAG-3') synthesized by Integrated DNA Technologies (Coralville, IA, USA). The Flag epitope was inserted between amino acids K191 and T192, located between transmembrane domains B and C in an extracellular loop of hCLC-3. Correct insertion of the Flag epitope was verified by sequencing. The Flag-CLC-3 construct was transfected into HT29 or tsA cells as described above.
Construction of mutant CLC-3
The serine at position 109 was mutated to an alanine (S109A) in hCLC-3 pcDNA3.1 and Flag-CLC-3, using the Quik Change Site-Directed Mutagenesis Kit (Stratagene). Escherichia coli transformed with mutant CLC-3 were grown at 37°C. The resultant S109A (± Flag) mutants were confirmed by sequence analysis. Mutant CLC-3 was cotransfected along with pEGFPN1 vector (Clontech) into tsA cells at a 10 : 1 ratio, using lipofectamine (Invitrogen). Transiently transfected cells were identified by their expression of green fluorescent protein (GFP). Experiments were performed at 48 h post-transfection.
Immunofluorescence
AIP/A23187 experiments.
Cells were grown on 25 mm coverslips coated with poly-D-lysine. Cells were incubated with PBS or 1 µM myristoylated autocamtide inhibitory peptide (Myr-AIP; Biomol Research Laboratories, Inc., PA, USA) for 30 min at room temperature, then with PBS or 10 µM A23187 (Sigma, MO, USA) at room temperature (RT) followed by fixation in 4% paraformaldehyde for 20 min at RT, and, if indicated, permeabilization with 0.1% Triton X-100 for 5 min at RT. Cells were incubated with -hCLC-3730–744 at a 1 : 500 dilution for 1 h at room temperature, then washed with PBS and incubated with AlexaFluor488 or rhodamine-conjugated goat -rabbit IgG (Molecular Probes, OR, USA) for 1 h at room temperature. Cells without exposure to the primary antibody or non-transfected cells were used as a control. The coverslips were mounted on a slide and observed using confocal microscopy (Olympus Fluoview).
CLC-3 localization experiments.
tsA cells were grown on 25 mm coverslips and the experimental group was transfected with 1 µg Flag-CLC-3 DNA. Cells were fixed in 4% paraformaldehyde for 20 min at RT, and depending on the experiment, permeabilized with 0.1% Triton X-100 for 5 min at RT. Cells were incubated with a mouse -Flag monoclonal antibody at a 1 : 1300 dilution (Sigma) or rabbit -hCLC-3730–744 at a 1 : 500 dilution for 1 h at room temperature, then washed with PBS and incubated with rhodamine or AlexaFluor488-conjugated goat -mouse IgG or AlexaFluor488-conjugated goat -rabbit IgG (Molecular Probes, OR, USA) for 1 h at room temperature. Cells without exposure to the primary antibody and non-transfected cells were used as controls. The coverslips were mounted on a slide and observed using confocal microscopy (Olympus Fluoview).
Electrophysiological studies
Whole-cell voltage clamp. Whole-cell patch clamp experiments were performed using isolated primary aortic smooth muscle cells, non-confluent HT29, or tsA cells. All electrophysiological methods were similar to those described earlier (Xie et al. 1996, 1998). Currents were elicited during a series of test pulses from –110 to +110 mV in 10 mV increments from a holding potential of –40 mV. Test pulses were 200 ms in duration and delivered at 2 s intervals. The pipette solution contained (mM): NMDG 140, HCl 40, L-glutamic acid 100, CaCl2 0.2, MgCl2 2, EGTA 1, Hepes 10 and ATP-Mg 2, pH 7.2. Free Ca2+ was 40 nM. The bath solution contained (mM): NMDG 140, HCl 140, CaCl2 2, MgCl2 1 and Hepes 10, pH 7.4. Purified rat brain CaMKII was dialysed daily against Pipes buffer (mM: Pipes 250, pH 7.0, EGTA 1, NaCl 150) using Slide-A-Lyser mini dialysis units, 7000 MWCO (Pierce, IL, USA). The autonomous CaMKII was prepared as previously described (Chan et al. 1994). The autonomous, autophosphorylated kinase was introduced into the cell via the pipette solution. A 100 µM stock solution of the membrane-permeant CaMKII specific inhibitor Myr-AIP was diluted to 1 µM with bath solution (Biomol Research Laboratories, Inc., PA, USA) (Ishida et al. 1995, 1998). Carbachol (Sigma) was diluted to 1 µM with bath solution and superfused onto the cells. Where indicated, cells were incubated for 10 min in bath solution +1 µM Myr-AIP (Biomol Research Laboratories, Inc.) prior to whole cell experiments.
Capacitance measurements
Whole-cell capacitance recordings were obtained using an EPC-9 computer-controlled patch clamp amplifier (HEKA Electronik, Lambrecht, Germany) running PULSE software (HEKA). The EPC-9 includes a built-in data acquisition interface (ITC-16, Instrutech, NY, USA). The software package controlled the stimulus and data acquisition for the software lock-in amplifier in the ‘sine + dc’ mode as described by Gillis (Gillis, 2000). The temporal resolution of the capacitance data was 40 ms per point using a 1 kHz, 20 mV sine wave. The holding potential in capacitance experiments was –5 mV to allow for a capacitive current recorded concurrently with a non-contaminating conductance recording.
All experiments were performed at room temperature. Data are expressed as means ±S.E.M. with the number of experiments in parentheses. The statistical significance of the results was assessed using Student's t test analysis.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Earlier studies from our laboratory have established that recombinant human CLC-3 over-expressed in a stable cell line is gated in a phosphorylation-dependent manner by exogenous CaMKII (Huang et al. 2001); therefore, we reasoned that the kinase-activated anion conductance in other cells might be due to the functional expression of CLC-3. Whole-cell patch clamp experiments were performed in order to determine the molecular identity of the CaMKII-activated Cl– current in CLC-3–/– and CLC-3+/+ mouse aorta smooth muscle cells. Experiments were performed in asymmetrical Cl– solutions, where the theoretical reversal potential was –31 mV. Under these conditions, non-specific leak current was identified as a depolarizing shift in the zero-current potential, and experiments in which the reversal potential was more depolarized than –15 mV were not included in the data analysis. Patch-clamp experiments were performed on single, non-confluent cells cultured for 1–2 days. Current elicited in the presence (ICl,CaMKII) or absence (ICl,Basal) of autonomous CaMKII in the pipette is shown in Fig. 1A. The autonomous CaMKII was introduced into the cell via the pipette as previously described (Chan et al. 1994; Kaetzel et al. 1994; Xie et al. 1996). Basal current recordings were taken following establishment of the whole-cell configuration prior to diffusion of the kinase into the cell. As the autonomous enzyme diffused into the cell, Cl– current reached a maximal current activation (at +110 mV) of 22.3 ± 3.3 pA pF–1 in CLC-3+/+ cells, and 3.3 ± 1.2 pA pF–1 in CLC-3–/– cells. Current in the presence of the kinase in CLC-3–/– cells was not different from basal current in either CLC-3–/– or CLC-3+/+ cells (1.7 ± 0.4 and 2.2 ± 0.8 pA pF–1, respectively). AIP is an effective inhibitor of the CaMKII-mediated current (Ishida & Fujisawa, 1994; Ishida et al. 1995). In CLC-3 expressing cells, CaMKII + AIP-activated Cl– current density (2.3 ± 0.5 pA pF–1, n= 5) was not significantly different from basal. The current–voltage (I–V) relationship for ICl,CaMKII in the cells was outwardly rectifying with a reversal potential of approximately –25 mV (Fig. 1B and E).
|
CLC-3 is phosphorylated in vitro by CaMKII at the N-terminus
CLC-3 has three putative CaMKII phosphorylation sites: two of these (S109, T713) are accessible from the internal side of the cell, whereas the intracellular accessibility to the third (S420) is not fully established (Fig. 2A). To assess whether CLC-3 could be phosphorylated and thereby possibly activated by CaMKII, in vitro phosphorylation experiments were performed. Two different methods were used: in one, CLC-3 was immunoprecipitated from whole-cell lysates, and in the other, GST fusion proteins were made with either the N- or C-terminus of CLC-3.
|
-hCLC-3730–744, phosphorylated with CaMKII in the presence of Ca2+, calmodulin, and -32P-ATP, and resolved on SDS-PAGE. The resulting autoradiograph shows gel bands detected at approximately 120 kDa, the predicted molecular mass for a glycosylated form of CLC-3 (Fig. 2B). Synapsin, a synaptic vesicle protein known to be phosphorylated in vivo by CaMKII was used as a control and was phosphorylated in this assay (84 kDa). To determine the region of CLC-3 involved in CaMKII phosphorylation, GST fusion proteins were made with either the N- or C-terminus of CLC-3. The GST-CLC-3 constructs were in vitro phosphorylated with CaMKII, as in the whole-cell lysate studies, and run out on SDS-PAGE. The resulting autoradiograph and Coomassie gel are shown in Fig. 2C. Note that the N-, but not the C-terminus of CLC-3 is phosphorylated by CaMKII. These data indicate that CLC-3 is phosphorylated by CaMKII, specifically in the N-terminal region of the protein containing the putative CaMKII phosphorylation site S109. CaMKII is not responsible for trafficking CLC-3 to the plasma membrane
Phosphorylation-dependent gating of the channel could be due to an increase in the insertion or trafficking of the channel to plasma membrane sites or could be due to a phosphorylation-dependent conformational change in the channel or associated regulatory subunit. In vitro phosphorylation experiments using the autonomous kinase demonstrated that the channel was phosphorylated directly (Fig. 2B).
To determine whether a change in cellular localization was associated with an increase in internal Ca2+, HT29 cells expressing endogenous CLC-3 or transfected with Flag-CLC-3 were treated with media (Control), a Ca2+ ionophore, A23187 (10 µM), or the calcium ionophore and the peptide CaMKII inhibitor, Myr-AIP (1 µM), prior to fixing the cells (Fig. 3A). AIP in conjunction with A23187 was used to show the relative contribution of CaMKII to changes in cellular redistribution of CLC-3. The control, non-transfected permeabilized cells showed extensive cytoplasmic staining, with a high perinuclear distribution of CLC-3. Although there was a considerable shift in the distribution of CLC-3 upon exposure of the cells to the ionophore from a perinuclear to a diffuse cytoplasmic distribution as can be seen in Fig. 3A, this Ca2+-dependent shift in immunofluorescence was not inhibited by the specific inhibitor, Myr-AIP, as seen in Fig. 3A. In Flag-CLC-3-transfected HT29 cells, there was no apparent Ca2+–dependent change in membrane expression of CLC-3 for any conditions, consistent with the hypothesis that CLC-3 is not dependent on CaMKII for trafficking to the plasma membrane.
|
To ensure that HT29 cells can be used as a model in which to study trafficking, the expression and function of CLC-3 in these cells were determined. Endogenous CLC-3 expressed in the human colonic tumour cell line HT29 was immunoprecipitated with -hCLC-3730–744, detected in an immunoblot with -hCLC-359–74, and compared to recombinant CLC-3 expressed stably in tsA cellsas shown in Fig. 3D. The immunoprecipitated (IP) channel protein in the HT29 cells had molecular mass of approximately 120 kDa, which corresponds to the glycosylated form of the protein. There was no band present in the HT29 supernatant (Sup).
As shown in Fig. 3E, HT29 cells exhibit a CaMKII-activated Cl– current. Basal current recordings were taken following establishment of the whole-cell configuration prior to diffusion of the kinase into the cell. As the autonomous enzyme diffused into the cell, Cl– current reached a maximum activation at 15.2 ± 1 min (n= 47) following initiation of the whole-cell configuration. The I–V relationship for ICl,CaMKII was outwardly rectifying with a reversal potential of approximately –20 mV (data not shown). Maximal current density at 110 mV was 27.7 ± 1.5 pA pF–1 (n= 47) in the presence of CaMKII, which is approximately 4 times the basal current when CaMKII was absent, 6 ± 0.7 pA pF–1 (n= 7). AIP was also used to establish the specificity of the CaMKII-induced current. There was no significant change in kinase-activated current amplitude over basal when 1 µM AIP was added to the pipette solution containing the kinase and the kinase-activating cocktail indicating that current activation in the presence of the autonomous kinase was due to a CaMKII-specific mediated phosphorylation event (Fig. 3E).
Mutation at amino acid S109 abolishes CaMKII-mediated Cl– current
As was shown earlier (Fig. 2), CLC-3 is phosphorylated in vitro on the N-terminal region of the channel protein, which includes only one putative CaMKII phosphorylation site, S109. To determine whether S109 contributes to the CaMKII-mediated activation of CLC-3, tsA cells were transfected with wild-type (wt) or S109A CLC-3. Cells expressing the mutation of serine to alanine at position 109 of CLC-3 (S109A CLC-3) did not show an increase over basal current density with the addition of autonomous CaMKII to the pipette solution (3.2 ± 0.8 pA pF–1, n= 4), whereas cells expressing wt CLC-3 demonstrated an increase in Cl– current following introduction of CaMKII into the cell (18 ± 1.3 pA pF–1, n= 8). Mock-transfected cells showed no difference in basal versus CaMKII conditions (Fig. 4). As it is possible that the lack of current activation with CaMKII in the S109A-transfected cells was due to a lack of membrane presence, cells transfected with Flag-containing CLC-3 were examined in immunostaining studies. In both the wt and S109A-transfected tsA cells, CLC-3 appears at the plasma membrane, as shown in non-permeabilized preparations (Fig. 4D). This indicates that the mutation does not cause mistrafficking of the channel to the membrane and is consistent with the hypothesis that S109 is important for CLC-3 activation by CaMKII.
|
Kinetic characteristics of CLC-3 activation
A representative time course of current activation in CLC-3 stably transfected tsA cells is shown in Fig. 4E. Following the calculations of Pusch & Neher (1988), the predicted time course of dialysis of CaMKII, a 600 kDa protein, into the cell is 19.2 min. The average time to peak current elicited by CaMKII in tsA cells is 18.5 ± 1.8 min (n= 9), a time course not different for other cell types tested (HT29: 15.2 ± 1 min, n= 47; aorta smooth muscle: 16.8 ± 0.8 min, n= 9, data not shown). Carbachol-elicited peak currents were observable over a shorter time course (HT29: 10.7 ± 4.2 min (n= 9); tsA: 11.9 ± 2.2 min (n= 8); smooth muscle cells: 13.1 ± 3.7 min (n= 11)) as compared to CaMKII (Fig. 4E). In addition to time course measurements, the time dependence of current activation and inactivation was evaluated in all three cell types studied. The majority of the cells (82.8%) examined did not exhibit time-dependent activation or inactivation under both CaMKII and carbachol stimulation (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-CLC-3 antibody directed to the intracellular C-terminus of the protein showed both cytoplasmic as well as plasma membrane distribution. Activation of the CaMKII-dependent conductance was not dependent upon an apparent translocation event as evidenced from immunostaining experiments and studies of changes in membrane capacitance in the presence of the autonomous kinase. We have shown that CLC-3 is phosphorylated in vitro by CaMKII, and that this phosphorylation occurs specifically at the N-terminus. Additionally, mutation of the putative CaMKII phosphorylation site in the N-terminus, S109, specifically abolished the CaMKII-activated Cl– conductance in transiently transfected tsA cells. We conclude therefore that when endogenous CLC-3 is expressed at plasma membrane sites it is gated by CaMKII-dependent phosphorylation at S109 in the N-terminus of the channel protein. Intracellular Cl– regulation mediated through changes in intracellular Ca2+ has been observed in many cell types, including epithelial cells (Petersen & Philpott, 1980; Petersen, 1992), neurones (Mayer, 1985; Scott et al. 1995; Frings et al. 2000), cardiac and smooth muscle cells (Pacaud et al. 1989a; Sorota, 1999). The molecular identity of the proteins underlying the ubiquitously expressed Ca2+-dependent anion conductances has remained elusive. The pharmacological properties of these Cl– conductances are as varied as the cell types in which they reside. These conductances are generally inhibited by DIDS and niflumic acid, but have varying responses to 5-nitro-2-(3-phenyl propylamino) benzoic acid (NPPB) (Sellinger et al. 1992; Tilly et al. 1992; Koumi et al. 1994; Morier & Sauvé, 1994; Eliassi et al. 1997; Breit et al. 1998). Ca2+-activated Cl– channels generally display an I > Br > Cl permeability sequence and are almost impermeable to glutamate (Evans & Marty, 1986). In addition, in some cell types, the Ca2+-activated Cl– conductance is decreased by specific inhibitors of CaMKII (Dascal et al. 1985; Wagner et al. 1991; Morris & Frizzell, 1993; Arreola et al. 1998), and increased by purified CaMKII (Nishimoto et al. 1991; Wagner et al. 1991; Ho et al. 2001; Huang et al. 2001), whereas others show no CaMKII regulation (Ishikawa, 1996; Arreola et al. 1998). Aside from their pharmacological profiles, these channels are diverse in their biophysical properties as well, with single channel conductances ranging from 1 to 70 pS, and rectification ranging from strong outward to relatively linear I–V relationships (Marty et al. 1984; Frizzell et al. 1986; Takahashi et al. 1987; Klockner, 1993; Koumi et al. 1994; Collier et al. 1996; Schlenker & Fitz, 1996; Nilius et al. 1997; Zdebik et al. 1997; Ho et al. 2001).
Since the identification of the Ca2+-activated chloride current (ICl,Ca) in a smooth muscle cell (Byrne & Large, 1987a,b), much has been learned about the pharmacological and biophysical characteristics of the combined Ca2+-activated Cl– conductances. Just as these conductances have similar biophysical and pharmacological profiles across smooth muscle types, they are also similar to ICl,Ca in non-smooth muscle cells. Ca2+-activated Cl– conductances in smooth muscle are physiologically important as mediators of membrane depolarization and contraction (Criddle et al. 1996; Lamb & Barna, 1998). These conductances are regulated by a variety of hormones and neurotransmitters and are gated through various Ca2+-mediated pathways, including Ca2+ influx, voltage-gated Ca2+ channels, and enzymes, including CaMKII (Amedee & Large, 1989; Pacaud et al. 1989b; Klockner & Isenberg, 1991). It has been shown previously that CaMKII-dependent phosphorylation variably activated Cl– conductances in smooth muscle preparations (Sumi et al. 1991; Leblanc & Lupien, 1997; Greenwood et al. 2000, 2001). Our studies extend these electrophysiological analyses and identify the CaMKII-activated Cl– channel in the aorta smooth muscle cells as plasma membrane resident CLC-3.
Previous studies of CLC-3 have reported that it is not a Ca2+-activated Cl– conductance (Weylandt et al. 2001; Arreola et al. 2002; Ogura et al. 2002; Jin et al. 2003). While our studies have shown that CLC-3 can be activated by a Ca2+-dependent pathway via carbachol, we demonstrate that CLC-3 activation is exclusively through a phosphorylation-dependent mechanism via CaMKII. Furthermore, our data concur with previous studies that have shown no difference in carbachol activation in CLC-3+/+ and CLC-3–/– cells. The absence of a decrease in carbachol-stimulated Cl– current in the knockout model could be attributed to an increase in other conductances acting in compensation for the lack of CLC-3.
Early reports of the cloning and functional expression of a Ca2+-activated tracheal epithelial chloride channel (CLCA) in oocytes suggested that the protein, which shares no sequence homology with CLC-3, might indeed be the channel responsible for salt and water regulation in cells lining the fluid-filled cavities of the body. This family of putative channels bears striking resemblance to lung endothelial cell adhesion molecules (Elble et al. 1997; Gandhi et al. 1998), exhibits a range of cellular expression throughout the body, and thus a variety of physiological functions. Members of the CLCA channel family are structurally distinct from the CLC channel family, with a hydropathy analysis prediction of four transmembrane domains. When expressed in heterologous expression systems, CLCA channels exhibit Ca2+-dependent gating, an event which does not appear to be phosphorylation-dependent and may be due to Ca2+ binding to the channel protein directly (Pauli et al. 2000; Thevenod, 2002; Ledoux et al. 2003). The carbachol-activated, AIP-insensitive current that we observed in our studies may be due to the activation of a member of the CLCA family endogenously expressed and activated directly following a transient rise in intracellular Ca2+.
Two bestrophins, proteins expressed in a variety of tissues, were recently cloned from Xenopus oocytes, which express high levels of Ca2+-activated Cl– channels. Expression of the bestrophins in HEK cells gave rise to a large Ca2+-activated Cl– conductance. The currents were voltage and time independent, and exhibited a permeability ratio of I– > Cl–. Bestrophin mutations produced non-functional channels that exert a dominant negative effect on wild-type channels, consistent with the hypothesis that bestrophins are the first molecularly identified Cl– channels that activate in a physiologically relevant range of intracellular [Ca2+] (Qu et al. 2003).
Sequence analysis predicts three possible CaMKII consensus sequences within CLC-3. Two of these sites are located on the N- and C-terminal domain and one within the J transmembrane segment (transmembrane nomenclature as in Dutzler et al. (2002). Thus, kinase-dependent channel activation could be due to phosphorylation of the channel protein or alternatively to the phosphorylation-dependent translocation of channels from the large cytoplasmic pool. While it is clear that a translocation event is not a prerequisite for channel activation in the HT29 cells, it is not clear from our studies that channel gating is identical in the cytoplasmic/endosomal versus the plasma membrane compartments. Channel gating could well be a function of subunit association. Although our in vitro phosphorylation study demonstrates that the channel is phosphorylated by the kinase, it is entirely possible that an associated regulatory subunit, which is tightly associated with the channel in the plasma membrane is, in fact, a substrate for CaMKII-dependent phosphorylation as well. Differential expression of the associated subunit between the cytoplasmic and plasma membrane compartments could account for the absence of constitutive channel gating observed in some heterologous expression studies (Friedrich et al. 1999).
It is clear that from our studies on Cl– uptake and acidification in mouse pancreatic ß-cells (Barg et al. 2001) as well as the studies of Li et al. in human hepatoma cells (Li et al. 2002) that CLC-3 is a functional intracellular channel. When expressed in the endosomal or secretory vesicle compartment within cells, the gating of CLC-3 does not appear to be dependent upon Ca2+-dependent phosphorylation and instead may be modulated by changes in intravesicular pH. It has been previously demonstrated that other intracellularly localized members of the CLC family are pH sensitive. CLC-4 is activated by external acidification in endogenously expressing skeletal muscle cells (Kawasaki et al. 1999), while in recombinant cells and oocytes, the acidification of external pH inhibited the CLC-4 conductance (Friedrich et al. 1999; Vanoye & George, 2002). CLC-5, another member of the CLC family closely related to CLC-3, was shown to be inhibited by external acidification in oocytes (Mo et al. 1999). The modulation of intravesicular channels by acidification may be a direct consequence of vacuolar H+-ATPase activity or quite possibly by a closely associated regulatory protein with unique cytoplasmic expression. We have proposed that CLC-3 activity might well be regulated by the mdr1-like 65 kDa protein present in the insulin granule membrane (Barg et al. 2001). The regulatory protein associated with CLC-3 might vary between cytoplasmic vesicle populations.
A further level of regulation may reside in variable channel stoichiometry dependent upon compartmental expression. Recent analysis of the crystal structure of a bacterial member of the CLC channel family reveals a unique dimeric architecture with the channel organized as a dimer with a functional pore within each monomer (Dutzler et al. 2002). Given that each monomer has a functional pore, it is entirely feasible that the channel may function as a monomer (perhaps subunit associated) in one compartment and as a dimer in another compartment. Thus, a difference in functional stoichiometry could account for differences in channel gating between plasma membrane and cytoplasmic sites.
Recently, a novel alternate splice variant of CLC-3 (CLC-3A) has been cloned, CLC-3B, that is expressed primarily in epithelial cells (Ogura et al. 2002). CLC-3B localizes to the Golgi where its association with cystic fibrosis transmembrane conductance regulator (CFTR)-interacting PDZ proteins promotes trafficking to the plasma membrane (Gentzsch et al. 2002; Ogura et al. 2002). The individual CLC-3 splice variant products cannot be differentiated with the antibodies used in these studies, therefore it is possible that both isoforms exist in smooth muscle and HT29 cells examined in electrophysiological and biochemical assays. Consequently, the data reported here could be a combination of properties of CLC-3A and CLC-3B, or possibly other, as yet unidentified, isoforms.
In summary, the recognition of CLC-3 as the CaMKII-activated Cl– conductance in aorta smooth muscle and colonic epithelia will provide further insights into the regulation of anion transport in the many other types of cells in which CLC-3 channels are expressed. Furthermore, the determination of a molecular species underlying a Ca2+-regulated anion conductance in a smooth muscle cell could allow insight into their regulation in other cell types and might provide new possibilities for pharmacological regulation of these channels in Cl– conductance diseases such as cystic fibrosis. The knowledge gained from these studies can now be used to determine various constituents of the signalling network and ascertain the precise molecular mechanisms that are involved.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Arreola J, Begenisich T, Nehrke K, Nguyen HV, Park K, Richardson L et al. (2002). Secretion and cell volume regulation by salivary acinar cells from mice lacking expression of the Clcn3 Cl– channel gene. J Physiol 545, 207–216.
Arreola J, Melvin J & Begenisich T (1998). Differences in regulation of Ca2+–activated Cl– channels in colonic and parotid secretory cells. Am J Physiol 274, C161–C166.[Medline]
Barg S, Huang P, Eliasson L, Nelson DJ, Obermuller S, Rorsman P et al. (2001). Priming of insulin granules for exocytosis by granular Cl– uptake and acidification. J Cell Sci 114, 2145–2154.
Breit S, Kolb H & Apfel H (1998). Regulation of ion channels in rat hepatocytes. Pflugers Arch 435, 203–210.[CrossRef][Medline]
Byrne N & Large W (1987a). Action of noradrenaline on single smooth muscle cells freshly dispersed from the rat anococcygeus muscle. J Physiol 389, 513–525.
Byrne N & Large W (1987b). Membrane mechanism associated with muscarinic receptor activation in single cells freshly dispersed from the rat anococcygeus muscle. Br J Pharmacol 92, 371–379.[Medline]
Chan HC, Kaetzel MA, Gotter AL, Dedman JR & Nelson DJ (1994). Annexin IV inhibits calmodulin-dependent protein kinase II-activated chloride conductance. J Biol Chem 269, 32464–32468.
Collier M, Levesque P, Kenyon J & Hume J (1996). Unitary Cl– channels activated by cytoplasmic Ca2+ in canine ventricular myocytes. Circ Res 78, 936–944.
Criddle D, de Moura RS, Greenwood I & Large W (1996). Effect of niflumic acid on noradrenaline-induced contractions of the rat aorta. Br J Pharmacol 118, 1065–1071.[Medline]
Dascal N, Gillo B & Lass Y (1985). Role of calcium mobilization in mediation of acetylcholine-evoked chloride currents in Xenopus laevis oocytes. J Physiol) 366, 299–313.
Dickerson LW, Bonthius DJ, Schutte BC, Yang B, Barna TJ, Bailey MC et al. (2002). Altered GABAergic function accompanies hippocampal degeneration in mice lacking CLC-3 voltage-gated chloride channels. Brain Res 958, 227–250.[CrossRef][Medline]
Duan D, Winter C, Cowley S, Hume JR & Horowitz B (1997). Molecular identification of a volume-regulated chloride channel. Nature 390, 417–421.[CrossRef][Medline]
Dutzler R, Campbell EB, Cadene M, Chait BT & MacKinnon R (2002). X-ray structure of a ClC chloride channel at 3.0 Å reveals the molecular basis of anion selectivity. Nature 415, 287–294.[CrossRef][Medline]
Elble RC, Widom J, Gruber AD, Abdel-Ghany M, Levine R, Goodwin A et al. (1997). Cloning and characterization of lung-endothelial cell adhesion molecule-1 suggest it is an endothelial chloride channel. J Biol Chem 272, 27853–27861.
Eliassi A, Garneau L, Roy F & Sauve R (1997). Characterization of a chloride-selective channel from rough endoplasmic reticulum membranes of rat hepatocytes: evidence for block by phosphate. J Membr Biol 159, 219–229.[CrossRef][Medline]
Erondu NE & Kennedy MB (1985). Regional distribution of type II Ca2+/calmodulin-dependent protein kinase in rat brain. J Neurosci 5, 3270–3277.[Abstract]
Evans MG & Marty A (1986). Calcium-dependent chloride currents in isolated cells from rat lacrimal glands. J Physiol 378, 437–460.
Freshney R (2000). Culture of Animal Cells: a Manual of Basic Technique. Wiley-Liss/John Wiley & Sons, Inc, New York.
Friedrich T, Breiderhoff T & Jentsch T (1999). Mutational analysis demonstrates that CLC-4 and CLC-5 directly mediate plasma membrane currents. J Biol Chem 274, 896–902.
Frings S, Reuter D & Kleene S (2000). Neuronal Ca2+-activated Cl– channels – homing in on an elusive channel species. Prog Neurobiol 60, 247–289.[CrossRef][Medline]
Frizzell RA, Rechkemmer G & Shoemaker RL (1986). Altered regulation of airway epithelial cell chloride channels in cystic fibrosis. Science 233, 558–560.
Gandhi R, Elble RC, Gruber AD, Schreur KD, Ji HL, Fuller CM et al. (1998). Molecular and functional characterization of a calcium-sensitive chloride channel from mouse lung. J Biol Chem 273, 32096–32101.
Gentzsch M, Cui L, Mengos A, Chang X, Chen J & Riordan J (2002). The PDZ-binding chloride channel CLC-3B localizes to the Golgi and associates with CFTR-interacting PDZ proteins. J Biol Chem 278, 6440–6449.
Gillis K (2000). Admittance-based measurement of membrane capacitance using the EPC-9 patch-clamp amplifier. Pflugers Arch 439, 655–664.[CrossRef][Medline]
Greenwood I, Ledoux J & Leblanc N (2000). Comparison of ICl (Ca) in arterial and vernous smooth muscle cells. Biophys J 78, 471A.
Greenwood I, Ledoux J & Leblanc N (2001). Differential regulation of Ca2+ activated Cl– currents in rabbit arterial and portal vein smooth muscle by Ca2+ calmodulin-dependent kinase. J Physiol 534, 395–408.
Gunther W, Luchow A, Cluzeaud F, Vandewalle A & Jentsch TJ (1998). CLC-5, the chloride channel mutated in Dent's disease, colocalizes with the proton pump in endocytically active kidney cells. Proc Natl Acad Sci U S A 95, 8075–8080.
Ho M, Kaetzel M, Armstrong D & Shears S (2001). Regulation of a human chloride channel. a paradigm for integrating input from calcium, type II calmodulin-dependent protein kinase, and inositol 3,4,5,6-tetrakisphosphate. J Biol Chem 276, 18673–18680.
Holevinsky KO, Jow F & Nelson DJ (1994). Elevation in intracellular calcium activates both chloride and proton currents in human macrophages. J Membr Biol 140, 13–30.[Medline]
Huang P, Di Liu JA, Robinson NC, Musch MW, Kaetzel MA et al. (2001). Regulation of human CLC-3 channels by multifunctional Ca2+/calmodulin- dependent protein kinase. J Biol Chem 276, 20093–20100.
Ishida A & Fujisawa H (1994). Stabilization of calmodulin-dependent protein kinase II through the autoinhibitory domain. J Biol Chem 270, 2163–2170.[CrossRef]
Ishida A, Kameshita I, Okuno S, Kitani T & Fujisawa H (1995). A novel highly specific and potent inhibitor of calmodulin-dependent protein kinase II. Biochem Biophys Res Commun 212, 806–812.[CrossRef][Medline]
Ishida A, Shigeri Y, Tatsu Y, Uegaki K, Kameshita I, Okuno S et al. (1998). Critical amino acid residues of AIP, a highly specific inhibitory peptide of calmodulin-dependent protein kinase II. FEBS Lett 427, 115–118.[CrossRef][Medline]
Ishikawa T (1996). A bicarbonate- and weak acid-permeable chloride conductance controlled by cytosolic Ca2+ and ATP in rat submandibular acinar cells. J Membr Biol 153, 147–159.[CrossRef][Medline]
Jentsch TJ (1996). Chloride channels: a molecular perspective. Curr Opin Neurobiol 6, 303–310.[CrossRef][Medline]
Jentsch TJ, Friedrich T, Schriever A & Yamada H (1999). The ClC chloride channel family. Pflugers Arch 437, 783–795.[CrossRef][Medline]
Jentsch TJ, Stein V, Weinreich F & Zdebik AA (2002). Molecular structure and physiological function of chloride channels. Physiol Rev 82, 503–568.
Jin N, Kim J, Yang D, Cho S, Kim J, Koh E et al. (2003). Fundamental role of CLC-3 in volume-sensitive Cl– channel function and cell volume regulation in AGS cells. Am J Physiol Gastrointest Liver Physiol 285, G938–G948.
Kaetzel MA, Chan HC, Dubinsky WP, Dedman JR & Nelson DJ (1994). A role for annexin IV in epithelial cell function. J Biol Chem 269, 5297–5302.
Kawasaki M, Fukuma T, Yamauchi K, Sakamoto H, Marumo F & Sasaki S (1999). Identification of an acid-activated Cl– channel from human skeletal muscles. Am J Physiol 277, C948–C954.[Medline]
Kawasaki M, Suzuki M, Uchida S, Sasaki S & Marumo F (1995). Stable and functional expression of the CIC-3 chloride channel in somatic cell lines. Neuron 14, 1285–1291.[CrossRef][Medline]
Kawasaki M, Uchida S, Monkawa T, Miyawaki A, Mikoshiba K, Marumo F et al. (1994). Cloning and expression of a protein kinase C-regulated chloride channel abundantly expressed in rat brain neuronal cells. Neuron 12, 597–604.[CrossRef][Medline]
Klockner U (1993). Intracellular calcium ions activate a low-conductance chloride channel in smooth-muscle cells isolated from human mesenteric artery. Pflugers Arch 424, 231–237.[CrossRef][Medline]
Klockner U & Isenberg G (1991). Endothelin depolarizes myocytes from porcine coronary and human mesenteric arteries through Ca-activated chloride current. Pflugers Arch 418, 168–175.[CrossRef][Medline]
Koumi S-I, Sato R & Aramaki T (1994). Characterization of the calcium-activated chloride channel in isolated guinea-pig hepatocytes. J General Physiol 104, 357–373.
Lamb F & Barna T (1998). Chloride ion currents contribute functionally to norepinephrine-induced vascular smooth muscle contraction. Am J Physiol 275, H151–H160.[Medline]
Leblanc N & Lupien M (1997). Regulation of Ca2+ dependent Cl– channels by calmodulin-dependent protein kinase in vascular myocytes. Biophys J 72, A33.
Ledoux J, Greenwood I, Villeneuve L & Leblanc N (2003). Modulation of Ca2+-dependent Cl– channels by calcineurin in rabbit coronary arterial myocytes. J Physiol 552, 701–714.
Li X, Shimada K, Showalter LA & Weinman SA (2000). Biophysical properties of CLC-3 differentiate it from swelling-activated chloride channels in Chinese hamster ovary-K1 cells. J Biol Chem 46, 35994–35998.
Li X, Wang T, Zhao Z & Weinman SA (2002). The CLC-3 chloride channel promotes acidification of lysosomes in CHO-K1 and Huh-7 cells. Am J Physiol Cell Physiol 282, C1483–C1491.
Marty A, Tan Y & Trautmann A (1984). Three types of calcium-dependent channels in rat lacrimal glands. J Physiol 357, 293–325.
Mayer M (1985). A calcium-activated chloride current generates the after-depolarization of rat sensory neurones in culture. J Physiol 364, 217–239.
Mo L, Hellmich H, Fong P, Wood T, Embesi J & Wills N (1999). Comparison of amphibian and human CLC-5: similarity of functional properties and inhibition by external pH. J Membr Biol 168, 253–264.[CrossRef][Medline]
Morier N & Sauvé R (1994). Analysis of a novel double-barreled anion channel from rat liver rough endoplasmic reticulum. Biophys J 67, 590–602.
Morris AP & Frizzell RA (1993). Ca2+-dependent Cl– channels in undifferentiated human colonic cells (HT-29). II. Regulation and rundown. Am J Physiol 264, C977–C985.[Medline]
Nilius B, Prenen J, Szucs G, Wei L, Tanzi F, Voets T et al. (1997). Calcium-activated chloride channels in bovine pulmonary artery endothelial cells. J Physiol 498, 381–396.[Medline]
Nishimoto I, Wagner JA, Schulman H & Gardner P (1991). Regulation of Cl– channels by multifunctional CaM kinase. Neuron 6, 547–555.[CrossRef][Medline]
Ogura T, Furukawa T, Toyozaki T, Yamada K, Zheng Y, Katayama Y et al. (2002). CLC-3B, a novel CLC-3 splicing variant that interacts with EBP50 and facilitates expression of CFTR-regulated ORCC. FASEB J 16, 863–865.
Pacaud P, Loirand G, Lavie J, Mironneau C & Mironneau J (1989a). Calcium-activated chloride current in rat vascular smooth muscle cells in short-term primary culture. Pflugers Arch 413, 629–636.[CrossRef][Medline]
Pacaud P, Loirand G, Mironneau C & Mironneau J (1989b). Noradrenaline activates a calcium-activated chloride conductance and increases the voltage-dependent calcium current in cultured single cells of rat portal vein. Br J Pharmacol 97, 139–146.[Medline]
Pauli BU, Abdel-Ghany M, Cheng HC, Gruber AD, Archibald HA & Elble RC (2000). Molecular characteristics and functional diversity of CLCA family members. Clin Exp Pharmacol Physiol 27, 901–905.[CrossRef][Medline]
Petersen OH (1992). Stimulus-secretion coupling: cytoplasmic calcium signals and the control of ion channels in exocrine acinar cells. J Physiol 448, 1–51.
Petersen OH & Philpott HG (1980). Mouse pancreatic acinar cells: the anion selectivity of the acetylcholine-opened chloride pathway. J Physiol 306, 481–492.
Pusch M & Neher E (1988). Rates of diffusional exchange between small cells
84 kDa. C, in vitro phosphorylation of GST CLC-3 fusion proteins by CaMKII. The GST CLC-3 or rabbit synapsin was phosphorylated with CaMKII in the presence of [
