Regulation of Human CLC-3 Channels by Multifunctional Ca2+/Calmodulin-dependent Protein Kinase
Ping Huang,Jie Liu,Anke Di,Nicole C. Robinson,Mark W. Musch,Marcia A. Kaetzel,Deborah J. Nelson
DOI: https://doi.org/10.1074/jbc.m009376200
2001-01-01
Abstract:The multifunctional calcium/calmodulin-dependent protein kinase II, CaMKII, has been shown to regulate chloride movement and cellular function in both excitable and non-excitable cells. We show that the plasma membrane expression of a member of the ClC family of Cl−channels, human CLC-3 (hCLC-3), a 90-kDa protein, is regulated by CaMKII. We cloned the full-length hCLC-3 gene from the human colonic tumor cell line T84, previously shown to express a CaMKII-activated Cl− conductance (ICl,CaMKII), and transfected this gene into the mammalian epithelial cell line tsA, which lacks endogenous expression of ICl,CaMKII. Biotinylation experiments demonstrated plasma membrane expression of hCLC-3 in the stably transfected cells. In whole cell patch clamp experiments, autonomously active CaMKII was introduced into tsA cells stably transfected with hCLC-3 via the patch pipette. Cells transfected with the hCLC-3 gene showed a 22-fold increase in current density over cells expressing the vector alone. Kinase-dependent current expression was abolished in the presence of the autocamtide-2-related inhibitory peptide, a specific inhibitor of CaMKII. A mutation of glycine 280 to glutamic acid in the conserved motif in the putative pore region of the channel changed anion selectivity from I− > Cl− to Cl− > I−. These results indicate that hCLC-3 encodes a Cl− channel that is regulated by CaMKII-dependent phosphorylation.AF172729 The multifunctional calcium/calmodulin-dependent protein kinase II, CaMKII, has been shown to regulate chloride movement and cellular function in both excitable and non-excitable cells. We show that the plasma membrane expression of a member of the ClC family of Cl−channels, human CLC-3 (hCLC-3), a 90-kDa protein, is regulated by CaMKII. We cloned the full-length hCLC-3 gene from the human colonic tumor cell line T84, previously shown to express a CaMKII-activated Cl− conductance (ICl,CaMKII), and transfected this gene into the mammalian epithelial cell line tsA, which lacks endogenous expression of ICl,CaMKII. Biotinylation experiments demonstrated plasma membrane expression of hCLC-3 in the stably transfected cells. In whole cell patch clamp experiments, autonomously active CaMKII was introduced into tsA cells stably transfected with hCLC-3 via the patch pipette. Cells transfected with the hCLC-3 gene showed a 22-fold increase in current density over cells expressing the vector alone. Kinase-dependent current expression was abolished in the presence of the autocamtide-2-related inhibitory peptide, a specific inhibitor of CaMKII. A mutation of glycine 280 to glutamic acid in the conserved motif in the putative pore region of the channel changed anion selectivity from I− > Cl− to Cl− > I−. These results indicate that hCLC-3 encodes a Cl− channel that is regulated by CaMKII-dependent phosphorylation.AF172729 Ca2+/calmodulin-dependent protein kinase II reverse transcriptase-polymerase chain reaction polyacrylamide gel electrophoresis phosphate-buffered saline 1,2-bis(O-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid 1,4-piperazinediethanesulfonic acid 4,4′-diisothiocyanostilbene-2,2′-disulfonic acid cystic fibrosis Gating of the chloride channel, CLC-3, which is expressed in brain and chloride (Cl−) secretory epithelial tissues has remained controversial since its initial identification and characterization (1Kawasaki M. Uchida S. Monkawa T. Miyawaki A. Mikoshiba K. Marumo F. Sasaki S. Neuron. 1994; 12: 597-604Abstract Full Text PDF PubMed Scopus (218) Google Scholar). The 760-amino acid protein encoded by theClc-3 gene was originally cloned from rat kidney and showed abundant expression in rat brain, most notably in the olfactory bulb, hippocampus, and cerebellum (1Kawasaki M. Uchida S. Monkawa T. Miyawaki A. Mikoshiba K. Marumo F. Sasaki S. Neuron. 1994; 12: 597-604Abstract Full Text PDF PubMed Scopus (218) Google Scholar). When expressed in a stably transfected cell line, the rat kidney isoform of the channel showed basal activation, inhibition by phorbol esters, and Ca2+sensitivity (2Kawasaki M. Suzuki M. Uchida S. Sasaki S. Marumo F. Neuron. 1995; 14: 1285-1291Abstract Full Text PDF PubMed Scopus (81) Google Scholar). Duan and colleagues (3Duan D. Winter C. Cowley S. Hume J.R. Horowitz B. Nature. 1997; 390: 417-421Crossref PubMed Scopus (413) Google Scholar, 4Duan D. Cowley S. Horowitz B. Hume J.R. J. Gen. Physiol. 1999; 113: 57-70Crossref PubMed Scopus (154) Google Scholar) characterized the functional expression of a cardiac clone of guinea pigclc-3, which when expressed in NIH 3T3 cells resulted in a large basally active Cl− conductance that was activated by an increase in cell volume, inhibited by phorbol esters and exhibited biophysical properties at the single channel level identical to the swelling activated current in native cardiac myocytes. Shimada and colleagues (5Shimada K. Li X. Xu G. Nowak D. Showalter L. Weinman S. Am. J. Physiol. Gastroint. Liver Physiol. 2000; 279: G268-G276Crossref PubMed Google Scholar) examined rat hepatocytes for Clc-3 expression and found that mRNA for two different isoforms was present; a short form corresponding to the guinea pig clone, and a long form containing a putative 58-amino acid addition at the N terminus of the protein. Both isoforms gave rise to current expression with identical selectivity when transiently expressed in CHO-K1 cells; however, they differed in the degree of outward rectification and voltage-dependent inactivation (5Shimada K. Li X. Xu G. Nowak D. Showalter L. Weinman S. Am. J. Physiol. Gastroint. Liver Physiol. 2000; 279: G268-G276Crossref PubMed Google Scholar). Adding a further dimension to the controversy, Friedrich and colleagues (6Friedrich T. Breiderhoff T. Jentsch T. J. Biol. Chem. 1999; 274: 896-902Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar) report in a mutational analysis of Clc-4 and Clc-5 that they were unable to detect currents upon Clc-3 expression in Xenopus oocytes or in transfected HEK293 cells. Recent evidence from the studies of Stobrawa et al. (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar) in transgenic mice with disrupted Clc-3 demonstrates that the chloride channel is broadly expressed and present in endosomal compartments and neuronal synaptic vesicles. Although theClc-3 knockout mice remained smaller than wild type littermates, they nonetheless remained viable for approximately 1 year. Among the most dramatic effects which Stobrawa and colleagues (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar) demonstrated in the Clcn-3−/− knockout mice were the nearly complete developmental loss of the hippocampus after birth as well as the complete loss of photoreceptors (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar). The loss of both the hippocampus and photoreceptors was attributed to inadequate acidification of synaptic vesicles. Defects observed in theClc-3-deficient mice lead Stobrawa and colleagues (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar) to hypothesize that Clc-3 acts as an anion shunt pathway which maintains charge balance for the proton pump which acidifies the vesicle interior prior to membrane fusion (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar). The finding that Clc-3 is an intracellular chloride channel expressed in both synaptic as well as endosomal vesicles suggests that it may be in dynamic equilibrium with the plasma membrane at a level which is dependent upon vesicular cycling in the various tissues in which it is expressed. Hence, expression of Clc-3 at surface sites will depend upon a balance between vesicle fusion with the plasma membrane and membrane retrieval. Regulated Ca2+-dependent Cl−conductances have been described in cells types as diverse as neurons, lymphocytes, and secretory epithelia. They regulate cellular volume, excitability, and salt balance. The multifunctional Ca2+/calmodulin-dependent protein kinase II (CaMKII)1 is a major mediator of Ca2+ signaling, is abundantly expressed in the brain, and is found in almost every cell type. Regulation of Cl−channels by CaMKII has been shown in cells from the human colonic tumor cell line, T84 (8Worrell R.T. Frizzell R.A. Am. J. Physiol. 1991; 260: C877-C882Crossref PubMed Google Scholar, 9Chan H.C. Kaetzel M.A. Gotter A.L. Dedman J.R. Nelson D.J. J. Biol. Chem. 1994; 269: 32464-32468Abstract Full Text PDF PubMed Google Scholar, 10Kaetzel M.A. Chan H.C. Dubinsky W.P. Dedman J.R. Nelson D.J. J. Biol. Chem. 1994; 269: 5297-5302Abstract Full Text PDF PubMed Google Scholar, 11Xie W. Kaetzel M.A. Bruzik K.S. Dedman J.R. Shears S.R. Nelson D.J. J. Biol. Chem. 1996; 271: 14092-14097Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar, 12Xie W. Solomons K.R.H. Freeman S. Kaetzel M.A. Bruzik K.S. Nelson D.J. Shears S.B. J. Physiol. 1998; 510: 661-673Crossref PubMed Scopus (49) Google Scholar, 13Merlin D. Jiang L. Strohmeier G.R. Nusrat A. Alper S.L. Lencer W.I. Madara J.L. Am. J. Physiol. 1998; 275: C484-C495Crossref PubMed Google Scholar), airway epithelia (14Wagner J.A. Cozens A.L. Schulman H. Gruenert D.C. Stryer L. Gardner P. Nature. 1991; 349: 793-796Crossref PubMed Scopus (154) Google Scholar), T lymphocytes (15Nishimoto I. Wagner J.A. Schulman H. Gardner P. Neuron. 1991; 6: 547-555Abstract Full Text PDF PubMed Scopus (62) Google Scholar), human macrophages (16Holevinsky K.O. Jow F. Nelson D.J. J. Membr. Biol. 1994; 140: 13-30Crossref PubMed Scopus (31) Google Scholar), biliary epithelial cells (17Schlenker T. Fitz J.G. Am. J. Physiol. 1996; 271: G304-G310Crossref PubMed Google Scholar), and cystic fibrosis-derived pancreatic epithelial cells (18Chao A.C. Kouyama K. Heist E.K. Dong Y.J. Gardner P. J. Clin. Invest. 1995; 96: 1794-1801Crossref PubMed Scopus (24) Google Scholar). The molecular identity of the channel or channels mediating the CaMKII-dependent conductance has remained unknown. Given the current controversy in the literature as to the activation pathway for heterologously expressed Clc-3, the ubiquitous expression of the CaMKII-activated Cl− conductance in native cells, and the presence of three possible CaMKII consensus sequences in the predicted amino acid sequence, our aim in the present study was to test the hypothesis that hCLC-3 is a Cl−channel regulated by CaMKII. For this study, we cloned a full-length hCLC-3 gene from T84 cells. This gene was 92% identical to the rat long form ofClc-3 reported by Shimada and colleagues (5Shimada K. Li X. Xu G. Nowak D. Showalter L. Weinman S. Am. J. Physiol. Gastroint. Liver Physiol. 2000; 279: G268-G276Crossref PubMed Google Scholar). Surface biotinylation experiments demonstrated that at least a portion of hCLC-3 is expressed at the surface of the stably transfected cells. Immunohistochemistry data showed that an increase in intracellular Ca2+ shifted the distribution of CLC-3 from a cytoplasmic, vesicular compartment to apparent surface sites. This translocation step was not required for current expression in the stably transfected cells as demonstrated in high resolution membrane capacitance measurements. We used the whole cell voltage-clamp technique to characterize the gating and selectivity of recombinant hCLC-3 channels stably expressed in a large-T antigen stabilized human embryonic kidney cell line HEK293 (tsA) cells. Our data show that the functional expression of the recombinant hCLC-3 induces Cl− conductance which is regulated by CaMKII with phenotypic properties of endogenous ICl,CaMKII in secretory epithelia. A mutation in the putative pore region, G280E, produced a characteristic change in anion selectivity from the I− > Br− > Cl− to Cl− > Br− > I− further demonstrating that the transfected channel was responsible for the current in the stably transfected cell line. Full-length hCLC-3 was cloned from a human colonic tumor cell line T84 by RT-PCR using SuperScript Preamplification System (Life Technologies Inc., MD). PCR primers (sense strand, TTGCTATGTCTCTGAGCTGC; antisense strand, AAGTAGATGACTCCCTCAGG) were derived from the hCLC-3 gene cloned from a human retina cDNA library by Borsani et al. (19Borsani G. Rugarli E.I. Taglialatela M. Wong C. Ballabio A. Genomics. 1995; 27: 131-141Crossref PubMed Scopus (77) Google Scholar). The PCR product was subcloned into pCR2.1-TOPO vector (Invitrogen) and sequenced. A comparison of the Borsani et al. (19Borsani G. Rugarli E.I. Taglialatela M. Wong C. Ballabio A. Genomics. 1995; 27: 131-141Crossref PubMed Scopus (77) Google Scholar) sequence with the T84 clone which we report reveals two variations. First, two amino acid residues (Glu647 and Phe658) are repeated in the Borsani et al. sequence. Second, a silent mutation at Ile772 (ATT) in the Borsani et al.(19Borsani G. Rugarli E.I. Taglialatela M. Wong C. Ballabio A. Genomics. 1995; 27: 131-141Crossref PubMed Scopus (77) Google Scholar) clone corresponds to Ile770 (ATC) in the T84 clone. In order to confirm the differences, we independently cloned the full-length hCLC-3 from tsA. The resultant sequence was consistent with the hCLC-3 cloned from T84 cells (GenBankTMAF172729). The full-length hCLC-3 was subcloned into the pcDNA3.1zeo+ vector (Invitrogen, CA) and transfected into tsA cells using SuperFect reagent (Qiagen, CA). Stably transfected clonal cell lines were selected using zeocin (Invitrogen) at 400 μg/ml and maintained at 200 μg/ml. The expression of hCLC-3 was detected by immunoblot analysis, and the homogenicity of the clone used in this study was confirmed ∼80% by immunostaining with hCLC-3-specific antibodies. The glycine at position 280 was mutated to a glutamic acid (G280E) using the QuikChangeTM Site-directed Mutagenesis Kit (Stratagene). Escherichia coli transformed with mutant hCLC-3 were grown at 30 °C. The resultant G280E mutation was confirmed by sequence analysis. Mutant hCLC-3 was co-transfected with pEGFPN1 vector (CLONTECH) into tsA cells at a 10:1 ratio, using SuperFect reagent (Qiagen). Transiently transfected cells were identified by their expression of enhanced green fluorescent protein. Whole cell electrophysiology recordings were performed at 48 h post-transfection. Antibodies were produced and affinity purified commercially by Quality Controlled Biochemical (Hopkinton, MA). Two synthetic peptides coupled to keyhole limpet hemocyanin were used to raise antibodies against hCLC-3 in rabbits. The peptides corresponded to amino acids 59–74 plus an additional cysteine (MTNGGSINSSTHLLDLC) and to amino acids 730–744 with an additional alanine and glutamic acid (GSSRVCFAQHTPSLPAE) in the C terminus. After a second boost, α-hCLC-3 serum was affinity purified using peptide coupled to thiol- or amino-linked gels. The stock concentration was 0.072 mg/ml for α-hCLC-359–74 and 1.03 mg/ml for α-hCLC-3730–744. The specificity of the antibodies was verified by immunoblotting whole cell lysates obtained from HEK293 stable cell lines expressing human CLC-1, rat Clc-2, and human CLC-4 (generous gifts of Drs. K. Blumenthal, A. George, and C. Fahlke). Cultured cells were harvested in H20E1D1 solution (20 mm HEPES, pH 8.0, 1 mm EDTA, 1 mmdithiothreitol) in the presence of a proteinase inhibitor mixture (Roche Molecular Biochemicals). The membrane and cytosolic fractions were crudely separated by centrifugation of the whole cell lysate at 200,000 × g for 30 min at 4 °C. Protein was denatured with 1/10 V of 10% SDS and 1 mm dithiothreitol, and heated to 95 °C for 5–8 min. For deglycosylation, the denatured protein (100 μg) was incubated with 5 units of N-glycanase F (Oxford GlycoSciences Ltd., Oxford, United Kingdom) for 18 h at 37 °C. Total protein (100 μg) was resolved by 7.5% SDS-PAGE and transferred to polyvinylidene difluoride membrane. Blots were incubated overnight at 4 °C with α-hCLC-359–74 at a 1:400 dilution or α-hCLC-3730–744 at a 1:3000 dilution, and subsequently with donkey α-rabbit antibody conjugated to horseradish peroxidase (Amersham Pharmacia Biotech) at a 1:2000 dilution for 20 min. Renaissance Western blot Chemiluminescence Reagent Plus (PerkinElmer Life Sciences) was used for detection. Following the protocol provided by Upstate Biotechnology (NY). Cells cultured on 100-mm dishes were lysed in 1 ml of modified RIPA buffer (in mm: Tris-HCl, 50; pH 7.4, NaCl, 150; and Nonidet P-40, 1%) containing proteinase inhibitors. One-third the volume of lysate was diluted with PBS to 1 ml, precipitated with 20 μg of α-hCLC-3730–744, then collected with 100 μl of 50% recombinant Protein A (Sigma). Immunoprecipitated protein was heated to 95 °C for 5–8 min in 1% SDS sample buffer containing 10 mm dithiothreitol and 100 mmβ-mercaptoethanol. One-fifth volume of the supernatant, prior to or following deglycosylation with 3 units of N-glycanase, was used for the Western blot. The membrane was immunoblotted with α-hCLC-359–74 at a 1:500 dilution. The tsA cells stably transfected with hCLC-3 were grown to ∼80% confluence in a 100-mm dish. The cells were washed three times with PBS, scraped, and pelleted (1,000 ×g, 2 min) into an Eppendorf microcentrifuge tube. Sulf-NHS-LC-LC-Biotin (Pierce) dissolved in PBS (pH 8.0, 0.5 mg/ml) was used to resuspend the cells at a concentration of 2.5 × 107 cells/ml. Incubation took place at 4 °C for 30 min on a rotating wheel. The cells were pelleted and the labeling step was repeated once. To terminate the biotinylation, 1/100 volume of 1m Tris-HCl, pH 7.5, was added and incubated for 3 min. Following incubation, the cells were washed with PBS three times for 2 min, and then immunoprecipitated in modified RIPA buffer with proteinase inhibitors. Total protein (300 μg) from the cell lysate was incubated with 40 μg of α-hCLC-3730–744 or PBS, pulled down with 100 μl of 50% Protein A (Sigma), and solublized in 60 μl of 2 × sample buffer containing 200 μmdithiothreitol. The resultant denatured protein was loaded for Western blot (30 μl/lane) analysis. The membrane was blotted with avidin-horesradish peroxidase conjugate (Bio-Rad) at a 1:5000 dilution and incubated overnight at 4 °C . Mucosal cells were harvested from ovine tracheal, rat ileal, and colonic epithelium. Brush-border was separated from basolateral membrane by differential centrifugation and precipitated with 15 mm CaCl2 as described by Bookstein et al. (20Bookstein C. dePaoli A.M. Xie Y. Niu P. Musch M.W. Rao M.C. Chang E.B. J. Clin. Invest. 1994; 93: 106-113Crossref PubMed Scopus (204) Google Scholar). Membrane protein (100 μg) prior to or following treatment with 5 units of N-glycanase was used for Western blot analysis. Cells were grown on 25-mm coverslips coated with poly-d-lysine. Cells were incubated with fresh medium or 10 μmA23187 (Sigma) for 5 min at room temperature followed by fixation in neutral-buffered 4% paraformaldehyde for 10 min at 4 °C, and permeabilization with 0.1% Triton X-100 for 3 min at 4 °C. Cells were incubated with either α-hCLC-3730–744 at a 1:400 dilution or mouse monoclonal anti-heat shock protein 27 (Affinity BioReagents, Inc.) at a 1:500 dilution for 1 h at room temperature, then washed with PBS, and incubated with AlexaFluor 488-conjugated goat α-rabbit IgG or goat α-mouse IgG (Molecular Probes, Eugene, OR) for 1 h at room temperature. Cells without exposure to the primary antibody were used as control. The coverslips were mounted on a slide and observed using confocal microscopy (Olympus Fluoview). Whole cell patch clamp experiments were performed on both T84 and hCLC-3 transfected tsA cells. All electrophysiological methods were similar to those described earlier (11Xie W. Kaetzel M.A. Bruzik K.S. Dedman J.R. Shears S.R. Nelson D.J. J. Biol. Chem. 1996; 271: 14092-14097Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar, 12Xie W. Solomons K.R.H. Freeman S. Kaetzel M.A. Bruzik K.S. Nelson D.J. Shears S.B. J. Physiol. 1998; 510: 661-673Crossref PubMed Scopus (49) Google Scholar). 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 (in mm): N-methyl D-glucamine, 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. In some of the experiments, 10 mm BAPTA was included in the pipette as the Ca2+ buffer in place of EGTA as indicated in the text. The bath solution contained (in mm): N-methyl D-glucamine, 140; HCl, 140; CaCl2, 2; MgCl2, 1; and HEPES, 10, pH 7.4. Purified rat brain CaMKII was dialyzed daily in PIPES buffer (in mm: PIPES, 25; pH 7.0, EGTA, 1; NaCl, 0.1) using Slide-A-LyzerTM Mini Dialysis Units, 7000 MWCO (Pierce). The autonomous CaMKII was prepared as previously described (9Chan H.C. Kaetzel M.A. Gotter A.L. Dedman J.R. Nelson D.J. J. Biol. Chem. 1994; 269: 32464-32468Abstract Full Text PDF PubMed Google Scholar). The autonomous, autophosphorylated kinase was introduced into the cell via the pipette solution. A 100 μm stock solution of the CaMKII-specific inhibitor autocamtide-2-related inhibitory peptide (21Ishida A. Kameshita I. Okuno S. Kitani T. Fujisawa H. Biochem. Biophys. Res. Commun. 1995; 212: 806-812Crossref PubMed Scopus (247) Google Scholar, 22Ishida A. Shigeri Y. Tatsu Y. Uegaki K. Kameshita I. Okuno S. Kitani T. Yumoto N. Fujisawa H. FEBS Lett. 1998; 427: 115-118Crossref PubMed Scopus (56) Google Scholar) in water was diluted to 1 μm with pipette solution (Biomol Research Laboratories, Inc.). The catalytic subunit of PKA (Promega) was diluted to 75 units/ml with pipette solution. 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). The software package controlled the stimulus and data acquisition for the software lock-in amplifier in the “sine + dc” mode as described by Gillis (23Gillis K. Pflügers Arch. Eur. J. Physiol. 2000; 439: 655-664Crossref PubMed Scopus (99) Google Scholar). The temporal resolution of the capacitance data was 40 ms per point using a 1 kHz, 20 mV sine wave. The holding potential in the capacitance experiments was −10 mV. All experiments were performed at room temperature. Data are expressed as mean ± S.E. with the number of experiments in parentheses. The statistical significance of the results was assessed using Student'st test analysis. The full-length hCLC-3 was cloned from the human colonic tumor cell line T84 using RT-PCR. The putative hCLC-3 protein is 818 amino acid residues in length sharing 92% identity with the long form Clc-3 cloned from rat hepatocytes (5Shimada K. Li X. Xu G. Nowak D. Showalter L. Weinman S. Am. J. Physiol. Gastroint. Liver Physiol. 2000; 279: G268-G276Crossref PubMed Google Scholar). The 58 residues at the N terminus are absent from guinea pig clc-3characterized by Duan et al. (3Duan D. Winter C. Cowley S. Hume J.R. Horowitz B. Nature. 1997; 390: 417-421Crossref PubMed Scopus (413) Google Scholar). The remaining 760-amino acid sequence shows 90% identity with both guinea pig and rat short form Clc-3. Analysis of the putative hCLC-3 amino acid sequence using CBS prediction servers (Center for Biological Sequence Analysis, Department of Biotechnology, The Technical University of Denmark) resulted in three putative CaMKII consensus sequences in hCLC-3 located at Ser109, Ser420, Thr713. We generated two polyclonal antibodies directed against peptides corresponding to the N- (α-hCLC-359–74) or C-terminal (α-hCLC-3730–744) sequences of hCLC-3. The specificity of the antibodies was verified by immunoblotting whole cell lysates obtained from HEK 293 stable cell lines expressing hCLC-1, rClc-2, and hCLC-4 (Fig.1 A). The corresponding predicted molecular mass for each of the chloride channel proteins was 107, 100, and 82 kDa, respectively. The affinity purified antibodies were used to evaluate expression of recombinant hCLC-3 in stably transfected tsA cells. Both antibodies detected a protein band from whole cell lysates with a molecular mass range from 90 to 120 kDa (Fig. 1 A). A single molecular mass band of ∼90 kDa was obtained from the same whole cell lysates digested with N-glycanase, which was consistent with the predicted mass for hCLC-3 of 88 kDa. The 90–120-kDa band was observed in the total membrane fraction, but absent from the cytosolic fraction. Given that CLC-3 has been localized to intracellular vesicles in cells from the central nervous system (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar), we carried out experiments to determine whether hCLC-3 was also expressed at the cell surface. Cells stably transfected with hCLC-3 were surface biotinylated, immunoprecipited with α-hCLC-3730–744 antibody or PBS, and subsequently immunoblotted with avidin-horesradish peroxidase conjugate. A single protein band was observed at approximately 90 kDa, and was absent in the PBS control (Fig. 1 B) clearly demonstrating that a significant fraction of hCLC-3 expression is at the cell surface. Expression of endogenous hCLC-3 in the human colonic epithelial cell line, T84, and in the murine macrophage cell line J774.1 was too low to be detected by standard Western blot analysis. However, protein expression in these cells was observed by immunoprecipitation with α-hCLC-3730–744 followed by immunoblot detection with α-hCLC-359–74. Following deglycosylation, endogenous hCLC-3 detected in both the gastrointestinal and immune cell lines had a molecular mass of 90 kDa (Fig. 2,IP). To determine protein localization in native, fluid-secreting tissues, the brush border and basolateral fractions of rat ileal and colonic as well as ovine tracheal epithelium were isolated by differential centrifugation. Membrane protein was resolved by 7.5% SDS-PAGE, and immunoblotted with α-hCLC-3730–744 antibody. An apparent molecular mass range from 90 to 130 kDa was detected in the brush border membrane fraction (Fig.3 A). The tissue dependent difference in apparent molecular weight for CLC-3 is consistent with Western analysis of membrane proteins from mice (7Stobrawa S. Breiderhoff T. Takamori S. Engel D. Schweizer M. Zdebik A. Boesl M. Ruether K. Jahn H. Draguhn A. Jahn R. Jentsch T. Neuron. 2001; 29: 185-196Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar). Tracheal CLC-3 revealed a high molecular mass band of greater than 120 kDa indicating an elevated level of glycosylation. CLC-3 detected in gastrointestinal tissue was characterized by a predominant molecular mass of 90 kDa, which did not shift after deglycosylation (Fig.3 B). Regulation of the subcellular distribution of CLC-3 was studied in T84 and the stably transfected tsA cells using confocal microscopy. Increases in the level of intracellular Ca2+ regulates vesicle trafficking and membrane fusion. We hypothesized that an increase in intracellular Ca2+ might shift the apparent expression of CLC-3 to surface sites. Cells were fixed and immunostained with α-hCLC-3730–744 either prior to or following exposure to the calcium ionophore, A23187 (10 μm). As seen in Fig. 4, the ionophore induced a translocation of fluorescence intensity from the cytoplasmic compartment to the plasma membrane in both cell lines. As a control, cells were immunostained under the same condition with another antibody targeting a non-secreted protein, heat shock protein 27 (HSP27). Fig. 5 demonstrates that HSP27 associated immunofluorescence remained in the cytoplasmic compartment before and after A23187 treatment.Figure 5Intracellular Ca2+ does not change subcellular localization of HSP27. T84 cells were immunostained with an antibody targeting a non-secreted protein, heat shock protein 27 (HSP27). Images were acquired with an Olympus Fluoview confocal microscope. HSP27 retained a diffuse cytoplasmic distribution prior to and following treatment with 10 μmA23187 for 5 min at room temperature. Bars, 10 μm.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Whole cell patch clamp experiments were carried out in order to examine the functional expression of hCLC-3. Experiments were performed in asymmetrical solutions where Cl− was the major permeant species, and the theoretical Cl− equilibrium potential was −31 mV. Under these ionic conditions, nonspecific leak current was identified as a depolarizing shift in zero-current potential. Voltage clamp experiments were performed on single, nonconfluent cells that had been maintained in culture for 1–2 days. The autonomous CaMKII was introduced into the cells via the patch pipette as has been previously described (9Chan H.C. Kaetzel M.A. Gotter A.L. Dedman J.R. Nelson D.J. J. Biol. Chem. 1994; 269: 32464-32468Abstract F