JBC Avanti Polar Lipids

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M400688200 on April 1, 2004

J. Biol. Chem., Vol. 279, Issue 23, 24673-24684, June 4, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/23/24673    most recent
M400688200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Li, C.
Right arrow Articles by Naren, A. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Li, C.
Right arrow Articles by Naren, A. P.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Molecular Assembly of Cystic Fibrosis Transmembrane Conductance Regulator in Plasma Membrane*

Chunying Li, Koushik Roy, Keanna Dandridge, and Anjaparavanda P. Naren{ddagger}

From the Department of Physiology, University of Tennessee Health Science Center, Memphis, Tennessee 38163

Received for publication, January 21, 2004 , and in revised form, March 31, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 REFERENCES
 
Based on electrophysiological measurements, it has been argued that the active form of cystic fibrosis trans-membrane conductance regulator (CFTR) Cl- channel is a multimer. It has also been demonstrated that this multimerization is likely due to PDZ domain-interacting partners. Here we demonstrate that although CFTR in vitro can self-associate into multimers, which depends on PDZ-based interactions, this may not be the case in cell membrane. Using chemical cross-linking, we demonstrated that CFTR exists as a higher order complex in cell membrane. However, this higher order complex is predominantly CFTR dimers, and the PDZ-interacting partners (Na+/H+ exchanger regulatory factor-1 (NHERF1) and NHERF2) constitute ~2% of this complex. Interestingly solubilizing membrane expressing CFTR in detergents such as Triton X-100, Nonidet P-40, deoxycholate, and SDS tended to destabilize the CFTR dimers and dissociate them into monomeric form. The dimerization of CFTR was tightly regulated by cAMP-dependent protein kinase-dependent phosphorylation and did not depend on the active form of the channel. In addition, the dimerization was not influenced by either the PDZ motif or its interacting partners (NHERF1 and NHERF2). We also demonstrated that other signaling-related proteins such as G{beta} and syntaxin 1A can be in this higher order complex of CFTR as well. Our studies provide a deeper understanding of how the CFTR assembly takes place in native cell membrane.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 REFERENCES
 
Many studies suggest that PDZ1 domain-containing proteins can cluster different proteins (such as ion channels, transporters, receptors, cytoskeletal proteins, and cytosolic signaling proteins) into functional complexes at synapses, cellular junctions, and polarized membrane microdomains (1). The cystic fibrosis transmembrane conductance regulator (CFTR) is a cAMP-activated Cl- channel localized primarily on the apical membrane of epithelial cells in the airway, gastrointestinal tract, reproductive organs, sweat duct, etc. where it is responsible for the transepithelial ion and fluid transport (2). An elegant functional study using electrophysiological measurements to identify microdomains under the apical membrane that transmit the signaling of CFTR via the adenosine receptor has been reported by Huang et al. (3) who have demonstrated that signaling elements compartmentalized at both extracellular and intracellular surfaces of apical cell membrane activate apical CFTR-dependent Cl- conductance in airway epithelial cells. We recently have shown that {beta}2 adrenergic receptor interacts with CFTR via Na+/H+ exchanger regulatory factor-1 (NHERF1, also called EBP50; Ref. 4) through their PDZ (PSD-95/Dlg/ZO-1) motifs, forming a macromolecular complex present at the apical surface of airway epithelial cells (5). Deleting the PDZ motif from CFTR uncouples the channel from {beta}2 adrenergic receptor both physically and functionally, and this uncoupling is specific to {beta}2 adrenergic receptor and does not affect CFTR coupling to other receptors (e.g. adenosine receptor pathway) (5). We further demonstrated that the assembly of the complex is regulated by cAMP-dependent protein kinase-dependent phosphorylation, and deleting the regulatory domain of CFTR abolishes cAMP-dependent protein kinase regulation of the complex assembly (5). Using co-immunoprecipitation and chemical cross-linking, Zhu et al. (6) reported that type 2C protein phosphatase is closely associated with CFTR in baby hamster kidney (BHK) cells stably expressing CFTR, suggesting that CFTR and type 2C protein phosphatase co-exist in a stable regulatory complex. The metabolic sensor AMP-activated protein kinase was found to interact directly with CFTR in vitro as well as in vivo (7, 8).

Controversy remains regarding the quaternary structure of CFTR Cl- channel at the apical surface of epithelial cells. Marshall et al. (9) concluded that CFTR exists primarily as a monomer in membrane since co-expressed full-length and C-terminal truncated forms of CFTR could not be co-immunoprecipitated in detergent-solubilized extract (Nonidet P-40, Triton X-100, digitonin, or CHAPSO). Using sucrose gradient centrifugation and co-expression of different epitope-tagged CFTR or wild-type and {Delta}F508 CFTR, Riordan's group (10) demonstrated that CFTR polypeptide was a monomer in detergent-solubilized extract as well (Nonidet P-40, Triton X-100, or SDS), while more and more compelling evidences suggest the existence of dimeric CFTR in plasma membrane (11-15). The cross-sectional area of the freeze-fracture particles corresponding to the area of the transmembrane domains of the recombinant CFTR in Xenopus oocytes showed that CFTR particle area corresponded more closely to 24 than to 12 packed helices, suggesting that CFTR channel may be dimeric in the membrane (11). A heterodimer consisting of a wild-type CFTR and a mutant CFTR ({Delta}R-CFTR) showed mixed gating properties of the wild-type and the mutant CFTR channels, implying that two CFTR molecules interact to form a single conductance pore for chloride ions (12). A study by Raghuram et al. (13) suggested that CFTR channel is a homodimer of CFTR molecules containing two PDZ domain binding sites whose gating is regulated by bivalent NHERF PDZ domains through cross-linking the C-terminal tails of the dimeric CFTR polypeptides. Recently, using gel filtration, non-dissociative gel electrophoresis, and chemical cross-linking, Ramjeesingh et al. (14, 15) reported that monomeric, dimeric, and multimeric CFTRs can be detected in Sf9 insect cells, and dimeric CFTR exists in the plasma membrane in mammalian cells.

In our present work, using a series of biochemical techniques, we focused on CFTR on native cell membrane. We found that, although CFTR can self-associate into multimers in a PDZ-based interaction-dependent manner, it exists as a dimer in native membrane and can be dissociated into a monomer upon solubilization of the membrane with detergents (e.g. Triton X-100 and Nonidet P-40). The PDZ-interacting partners NHERF1 and NHERF2 (also called E3KARP, SIP-1, or TKA-1; Ref. 16) contribute ~2% to this CFTR dimer-dominant complex, and this dimerization is PDZ motif-independent and channel activity-insensitive but cAMP-dependent protein kinase phosphorylation-dependent. Our study also showed that other proteins (such as G-protein {beta} subunit (G{beta}) and syntaxin 1A) can be associated with CFTR and may function in an integrated and regulatory pattern in the highly compartmentalized microdomains in the apical surface of epithelial cells.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
 REFERENCES
 
Cell Culture and Transfection—Calu-3 cells (serous gland epithelial cells) and COS-7 cells (African green monkey kidney cells) were obtained from American Type Culture Collection (Manassas, VA). Calu-3 cells were cultured as described previously (17). COS-7 cells and HT29-CL19A cells (human colonic epithelial origin) were grown in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal calf serum. BHK cells stably expressing CFTR (BHK-CFTR) and C-terminal polyhistidine-tagged CFTR (BHK-CFTRHis10) were grown as described previously (6). Cells were transiently transfected using LipofectAMINE 2000 (Invitrogen) as described previously (5) according to the manufacturer's instruction (for transfection of {beta}2 adrenergic receptor, NHERF1, {Delta}TRL-CFTR, and HA-tagged CFTR). BHK-CFTR or BHK-CFTRHis10 cells were transiently transfected with syntaxin 1A {Delta}C (deletion of the C-terminal membrane anchor) using a vaccinia virus expression system (18).

Antibodies and Reagents—CFTR monoclonal antibodies R1104 mAb recognizing the regulatory domain of CFTR (epitope at amino acids 722-734), GA1 mAb (amino acids 1440-1460), and NBD-R (amino acids 521-828) polyclonal antibody have been described previously (19). Anti-CFTR IgG MM13-4 was obtained from Chemicon (Temecula, CA). Anti-NHERF1, inositol 1,4,5-trisphosphate receptor, and BIR repeat containing ubiquitin-conjugating enzyme antibodies were obtained from BD Transduction Laboratories (Lexington, KY). Affinity-purified anti-syntaxin 1A IgG (14D8) has been described previously (20). Affinity-purified rabbit anti-G{beta} IgG was from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-NHERF2 antibody was a kind gift from Dr. Emanuel Strehler (Mayo Clinic, Rochester, MN; Ref. 21). HA-tagged CFTR cDNA (pIRES2-EGFP-HA-CFTR) was a kind gift from Dr. David Gadsby (Rockefeller University). cpt-cAMP, forskolin, and 3-isobutyl-1-methyl-xanthine were from ICN Biomedicals Inc. (Irvine, CA). N-[2-((p-Bromocinnamyl)amino)ethyl]-5-isoquinolinesulfonamide, 2HCl (H89) and diphenylamine-2-carboxylate (DPC) were from Calbiochem. Biotin-XX-hydrazide, sulfo-NHS-LC-biotin, and cross-linkers were obtained from Pierce. Cross-linkers used in the present study were: dithiobis(succinimidyl propionate) (DSP; spacer arm, 12.0 Å; membrane-permeable, cleavable by thiols), 1,5-difluoro-2,4-dinitrobenzene (DFDNB; spacer arm, 3.0 Å; membrane-permeable, not cleavable by thiols), disuccinimidyl tartarate (spacer arm, 6.4 Å; membrane-permeable, cleavable by periodate), dimethyl 3,3'-dithiobis(propionimidate) (DTBP; spacer arm, 11.9 Å; membrane-impermeable; cleavable by thiols), 3,3'-dithiobis(sulfosuccinimidyl propionate) (DTSSP; spacer arm, 12.0 Å; membrane-impermeable; cleavable by thiols), and ethylene glycol bis(succinimidyl succinate) (EGS; spacer arm, 16.1 Å; membrane-permeable; cleavable by hydroxylamine).

CFTR Multimer Assembly—This assay was performed by using maltose-binding protein (MBP)-CFTR-C-tail fusion protein (0-1 µM) immobilized on amylose beads (20 µl) and incubated with NHERF1 or NHERF2 (either with glutathione S-transferase (GST) tag or cleavage of the GST portion with thrombin). This step, which is called pairwise binding, was done in 200 µl of lysis buffer (PBS, 0.2% Triton X-100 supplemented with a mixture of protease inhibitors containing 1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µg/ml pepstatin) and mixed at 22 °C for 60 min. For pairwise binding, the protein was eluted with Laemmli sample buffer and subjected to Western blot using anti-NHERF1 or anti-NHERF2 antibodies. For multimer assembly assay, the complex was washed once with the same buffer and allowed to bind CFTR or CFTRHis10 from BHK cell lysates. The binding was done at 4 °C for 3 h with constant mixing. The complex was washed twice with lysis buffer, and the amylose beads were eluted with sample buffer and subjected to Western blotting using R1104 mouse monoclonal CFTR antibody.

Pull-down Assay—The pull-down assay was performed as described before (19). Briefly cells were lysed with PBS, 0.2% Triton X-100 (unless otherwise stated) supplemented with protease inhibitors. After 16,000 x g centrifugation of cell lysates, and GST or GST fusion proteins (e.g. NHERF1, NHERF2, or syntaxin 1A {Delta}C) were added to aliquots of the supernatants. 20 µl of glutathione-agarose beads (50% slurry in H2O) were added after mixing for 30 min at 4 °C. The mixture was rotated for another 2 h at 4 °C. After washing three times with the same lysis buffer, the proteins were eluted from beads with sample buffer (containing 2.5% {beta}-mercaptoethanol). Eluates were separated on a 4-15% gel and immunoblotted with anti-CFTR (R1104), anti-G{beta}, or anti-syntaxin 1A (14D8) antibodies.

Chemical Cross-linking of CFTR Polypeptides—Cells (Calu-3 or BHK) were washed three times with prewarmed (37 °C) PBS supplemented with 0.1 mM CaCl2 and 1 mM MgCl2 (PBS-C/M). The cells were incubated with prewarmed, freshly made cross-linkers in PBS-C/M at 37 °C. Stocks were freshly prepared in Me2SO for cell-permeable cross-linkers and in PBS-C/M for cell-impermeable cross-linkers. Control cells were treated with prewarmed PBS-C/M (no cross-linkers). The reaction was stopped by incubating cells with chilled Tris-buffered saline (pH 7.5) for 1 min on ice. Cells were washed with PBS-C/M twice for 2 min each and then lysed with RIPA buffer (150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl, pH 7.4) or PBS, 0.2% Triton X-100 supplemented with the protease inhibitors. For quantitation of CFTR and NHERF1 protein levels, Calu-3 cells were cross-linked with 1 mM DSP as above, lysed in RIPA buffer, and subjected to co-immunoprecipitation with anti-NBD-R IgG. The immunoprecipitated complex was treated with 2.5% {beta}-mercaptoethanol ({beta}ME) to dissociate the cross-linked complex and resolved by 10% SDS-PAGE. Purified CFTRHis10 was used as a standard to quantitate CFTR, and GST-NHERF1-PDZ2 was used as a standard to quantitate the co-immunoprecipitated NHERF1 as described previously (20).

Immunoblotting and Co-immunoprecipitation—Cells were cross-linked with DSP or DFDNB for 10 min at 37 °C and then lysed with RIPA buffer (+protease inhibitors) on ice for 20 min. The cell lysates were rocked for 30 min at 4 °C, and the insoluble material was removed by centrifugation at 16,000 x g for 15 min at 4 °C. Protein concentration of the cell lysates was determined by the bicinchoninic assay (Pierce). For co-immunoprecipitation, CFTR polyclonal antibody (anti-NBD-R IgG, 1.0 µg) was cross-linked to 20 µl of protein A/G-agarose (Santa Cruz Biotechnology) as described before (5). The supernatant was incubated and mixed with the cross-linked beads overnight at 4 °C. The beads were washed three times with RIPA buffer before the immunoprecipitated proteins were eluted with sample buffer containing 2.5% {beta}-ME. The immunoprecipitated proteins were then separated on a 4-15% gel and blotted using monoclonal NHERF1 antibody. For immunoblotting, cell lysates were separated by SDS-PAGE (5% gel for cross-linking experiments, 4-15% gel for other experiments), transferred to polyvinylidene difluoride membranes, and immunoblotted for CFTR, NHERF1, NHERF2, G{beta}, syntaxin 1A, or HA tag using specific antibodies in combination with HRP-conjugated anti-mouse IgG or HRP-conjugated anti-rabbit IgG antibodies. The immunoreactive bands were visualized by ECL (Amersham Biosciences).

Colocalization and Quantitation of CFTR and NHERF1—Fully polarized monolayers of Calu-3 cells were cross-linked with 1.0 mM DSP at 37 °C for 10 min. Next the cells were biotinylated apically by using 0.5 mg/ml sulfo-NHS-LC-biotin at 4 °C for 30 min as described before (5). Monolayers from 12 permeable supports (24-mm diameter) were scraped off and homogenized, and the plasma membrane was separated on a sucrose cushion by ultracentrifugation as described below (see "Preparation of Plasma Membrane and Other Subcellular Fractions"). Streptavidin-agarose beads (100 µl) were used to capture the biotinylated membranes, and the membranes were subjected to immunoprecipitation using anti-NBD-R IgG. The immunoprecipitated complex was treated with 2.5% {beta}ME to dissociate the cross-linked complex and resolved be 10% SDS-PAGE. The relative amount of CFTR and NHERF1 in the biotinylated plasma membrane was estimated using purified CFTRHis10 and GST-NHERF1-PDZ2 as standards as described above.

Purification of Cross-linked His-tagged CFTR and HA-tagged CFTR—BHK-CFTRHis10 cells were incubated with 1 mM DSP for 10 min at 37 °C and lysed in RIPA buffer. CFTRHis10 was purified using 50 µl of Talon beads (BD Biosciences) for 30 min at 4 °C and washed three times with the same buffer. The protein was eluted from beads with 100 mM imidazole containing 0.2% Triton X-100 and subjected to 5% SDS-PAGE followed by immunoblotting using R1104 antibody. For purification of co-expressed His-tagged and HA-tagged CFTR, BHK-CFTRHis10 and BHK parental cells were transiently transfected with HA-tagged CFTR as described previously (5). The cells were lysed in RIPA buffer, purified using Talon beads, and eluted with imidazole as above. The eluted protein was resolved on a 5% gel and immunoblotted with anti-HA tag antibody.

Cross-linking and Biotinylation of Cell Surface CFTR—Calu-3 cells were cross-linked with 1 mM DSP or DFDNB as described above. Cell surface biotinylation of glycoproteins was performed as described previously (22). Briefly, after cross-linking, the cells were incubated with 10 mM sodium m-periodate in PBS-C/M at 4 °C in the dark for 60 min. After two washings with PBS-C/M and one wash with 100 mM sodium acetate containing 0.1 mM CaCl2 and 1 mM MgCl2 (NaAc-C/M, pH 5.5), cells were surface-biotinylated with 0.5 mg/ml biotin-XX-hydrazide (in NaAc-C/M) at 4 °C in the dark for another 60 min. The cells were rinsed twice with PBS-C/M and lysed in RIPA buffer. CFTR was immunoprecipitated using anti-NBD-R IgG and subjected to 5% SDS-PAGE. The blot was first probed using streptavidin-HRP (0.2 µg/ml), and the same blot was stripped and reprobed for CFTR (R1104).

Velocity Gradient Centrifugation—Sucrose gradient sedimentation was performed according to published protocols (10). Briefly Calu-3 cells were incubated with or without prewarmed 1 mM DSP or DFDNB in PBS-C/M at 37 °C for 10 min. Cells were scraped into homogenization buffer (250 mM sucrose, 20 mM HEPES, 1 mM EDTA, pH 7.4, supplemented with a protease inhibitor mixture) and then subjected to 20 strokes in a tight fitting Pyrex homogenizer. Homogenates were centrifuged first at 6500 x g for 5 min, and microsomal membranes were then pelleted by centrifugation of the postnuclear supernatant at 100,000 x g for 20 min. Membranes were solubilized in 1 ml of PBS, 0.2% Triton X-100 or RIPA buffer supplemented with protease inhibitors and mixed for 10 min at 4 °C. 1 ml of membrane extract was layered on top of an 11-ml linear, continuous 10-36% sucrose gradient containing the respective detergent buffers. The sucrose gradient was formed by a gradient former (Bio-Rad) and layered from the bottom to the top using a peristaltic pump (Amersham Biosciences). Following centrifugation at 260,000 x g in an SW41 Ti rotor for 12 h, 24 500-µl fractions were collected from the bottom, resolved by SDS-PAGE, and blotted with CFTR monoclonal antibody R1104. The sedimentation of molecular mass standards (Bio-Rad) containing thyroglobulin (670 kDa), bovine {gamma} globulin (158 kDa), chicken ovalbumin (44 kDa), and equine myoglobin (17 kDa) was performed as above after dilution in 1 ml of PBS, 0.2% Triton X-100 or RIPA buffer. 24 500-µl fractions were collected, resolved by 4-15% SDS-PAGE, and stained with GelCode Blue Stain reagent (Pierce). The sucrose concentration of each fraction was determined by a handheld 0-10° Brix refractometer (Fisher) after a 1:3 sample dilution.

Preparation of Plasma Membrane and Other Subcellular Fractions—Plasma membrane and other subcellular fractions were prepared according to methods described previously (23). In brief, cells were homogenized in buffer A (20 mM HEPES, pH 7.4, 1 mM EDTA, 250 mM sucrose, 1 mM dithiothreitol, and protease inhibitors) and centrifuged at 16,000 x g for 15 min. The pellet was suspended in buffer B (20 mM HEPES, pH 7.4, 1 mM EDTA, and protease inhibitors), applied on a 1.16 M sucrose cushion in buffer B, and centrifuged at 100,000 x g for 1 h. Plasma membrane appeared as a white band on top of the cushion that was diluted in buffer B and pelleted at 16,000 x g and resuspended in buffer A. Supernatant of post-16,000 x g homogenate was centrifuged at 40,000 x g, and the resulting supernatant was further centrifuged at 200,000 x g. The pellet was suspended in buffer A, and the 200,000 x g supernatant was considered the cytosolic fraction.

Short Circuit Current Measurements—Calu-3 and HT29-CL19A polarized cell monolayers were grown to confluency on Costar Transwell permeable supports. Filters were cross-linked with 0.5 mM DSP and then mounted in an Ussing chamber, and short circuit current measurement was carried out as described previously (24). Epithelia were bathed in Ringer's solution (basolateral: 140 mM NaCl, 5 mM KCl, 0.36 mM K2HPO4, 0.44 mM KH2PO4, 1.3 mM CaCl2, 0.5 mM MgCl2, 4.2 mM NaHCO3, 10 mM HEPES, 10 mM glucose, pH 7.2, [Cl-] = 149 mM), and low Cl- Ringer's solution (apical: 133.3 mM sodium gluconate, 5 mM potassium gluconate, 2.5 mM NaCl, 0.36 mM K2HPO4, 0.44 mM KH2PO4, 5.7 mM CaCl2, 0.5 mM MgCl2, 4.2 mM NaHCO3, 10 mM Hepes, 10 mM mannitol, pH 7.2, [Cl-] = 14.8 mM) at 37 °C, gassed with 95% O2 and 5% CO2.

Two-dimensional Gel Electrophoresis—Two-dimensional gel electrophoresis was performed according to the manufacturer's instruction (Bio-Rad). Briefly Calu-3 cells were cross-linked with 1 mM DSP as above followed by lysis with RIPA buffer. CFTR was immunoprecipitated with mouse anti-CFTR antibody (MM13-4) or normal mouse IgG immobilized on Protein A/G PLUS-agarose beads (Santa Cruz Biotechnology) through constant mixing at 4 °C overnight. 7-cm immobilized pH gradient strips were rehydrated with 150 µl of CFTR proteins extracted with Sequential Extraction Reagent II (8 M urea, 4% CHAPS, 40 mM Tris, 0.2% Bio-Lyte 3/10) supplemented with 2 mM tri-butyl phosphine at 20 °C for 12-16 h. Isoelectric focusing was performed at 20 °C in an isoelectric focusing tray using the following program: 250 V for 15 min, linear ramp, S3 = 4000 V until 20,000 V-h were reached. The strips were equilibrated for 10 min in a solution containing 6 M urea, 3% SDS, 375 mM Tris-HCl (pH 8.6), 30% (v/v) glycerol, 2% (w/v) dithiothreitol. A second equilibration step was carried out for another 10 min in the same solution containing 3% (w/v) iodoacetamide instead of dithiothreitol. The proteins in the strip were separated in a 10% polyacrylamide gel and immunoblotted with affinity-purified anti-G{beta} IgG or anti-syntaxin 1A IgG (14D8). A molecular marker gel was stained with GelCode Blue Stain reagent (Pierce), and images were acquired and analyzed with Quantity One (Bio-Rad).

Other Methods and Data Analysis—Iodide efflux was performed as described before (25). Results are given as mean ± S.E. for the indicated number of experiments. Curve generation was carried out using Origin software (Microcal Software, Inc., Northampton, MA).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
REFERENCES
 
CFTR Can Form a PDZ Domain-dependent Multimer in Vitro—Several groups using electrophysiological techniques have suggested that the active form of CFTR Cl- channel is a multimer and that multimerization is due largely to PDZ domain-interacting proteins (13, 26). We developed an in vitro biochemical assay to assemble the CFTR multimer complex, which is pictorially presented in Fig. 1A. First a complex between MBP-CFTR-C-tail (immobilized on amylose beads) and GST-NHERF1 or NHERF2 was formed in lysis buffer (PBS, 0.2% Triton X-100) demonstrated by pairwise binding that detects interaction between two individually purified proteins as shown in Fig. 1B (19), which is consistent with previously published results showing that both NHERF1 and NHERF2 can interact with CFTR (27). After washing this complex once with lysis buffer, BHK cell lysates expressing recombinant CFTR or CFTRHis10 (CFTRHis10 contains 10 His residues in the C-terminal tail; this protein has an internalized PDZ motif and cannot bind NHERF, Ref. 5) were added to the mixture and allowed to bind at 4 °C for 3 h. The eluted protein complex was probed by immunoblotting for CFTR (R1104). As shown in Fig. 1C, we detected a multimeric complex containing the C terminus of CFTR, NHERF1 or NHERF2, and full-length CFTR in vitro (Fig. 1C). GST did not form such multimers and was used as negative control. Also CFTRHis10 lacking the C-terminal PDZ motif could not form a complex (Fig. 1C). The inputs are shown as well (Fig. 1C). This complex formation increased linearly with increasing amounts of NHERF1 (Fig. 1D). Similar results also were observed using NHERF2 (data not shown). All of the in vitro binding studies were performed under non-saturating conditions.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 1.
CFTR can form a PDZ domain-dependent multimer in vitro. A, the pictorial representation of the in vitro CFTR multimer assembly (see "Experimental Procedures"). B, C-terminal tail of CFTR interacted directly with NHERF1 or -2 as determined by pairwise binding assay (19). A complex between MBP-CFTR-C-tail (0.5 µM) and GST-NHERF1 or -2 (0.5 µM) was formed in lysis buffer (200 µl final volume) as described under "Experimental Procedures." The protein was eluted with sample buffer and subjected to Western blot using anti-NHERF1 (left panel) or anti-NHERF2 (right panel) antibodies. C, multimer complex of CFTR-C-tail, NHERF1 or -2, and full-length CFTR was detected with BHK-CFTR cells but not with BHK-CFTRHis10 cells. See "Experimental Procedures" for details. D, dose-dependent multimer complex formation using BHK-CFTR cell lysates with increasing amounts of NHERF1. Saturation was not observed up to 3.3 µM.

 
CFTR Can Exist as a Higher Order Complex in Plasma Membrane—Given that CFTR can form multimers in vitro through PDZ motif-dependent protein-protein interactions, it is of interest to determine whether such multimers exist in the plasma membrane of cells endogenously expressing CFTR as well. Toward this goal, we performed chemical cross-linking of proteins. Protein cross-linking is a widely used method of determining near neighbor relationships of proteins and molecular associations in cell membranes (28-30). We applied various reagents to cross-link CFTR polypeptides in Calu-3 cells at 0.5 mM for 10 min at 37 °C (see "Experimental Procedures" for details). Protein lysates (50 µg) were subjected to Western blotting analysis and probed with R1104 CFTR antibody. Using at least two different cross-linkers (DSP and DFDNB, both of which are membrane-permeable), we detected the existence of a higher order CFTR-containing complex (Fig. 2A, upper band). We also noted that DFDNB is only 3 Å in arm length, suggesting that the cross-linked components in this complex are probably in relatively close proximity. Although DSP and DFDNB had the most efficient effect, we were able to demonstrate cross-linking using other cross-linkers (e.g. EGS or disuccinimidyl tartarate) at 0.5-10 mM concentration range over 30-60-min time points (data not shown). Also we observed cross-linking by using cell-impermeable cross-linkers (DTBP or DTSSP) at 10 mM for 30 min (data not shown). DSP has a cleavable disulfide bond (31) and can be disrupted using 2.5% {beta}ME (Fig. 2B), while the DFDNB-dependent multimer cannot be cleaved (data not shown). Using cross-linkers, we wanted to quantitate the amount of NHERF present in the cross-linked complex with CFTR. For these experiments, we used RIPA buffer. The rationale was as follows. First, low affinity protein-protein interactions between CFTR and NHERF1 are disrupted by this buffer as 0.1% SDS and 0.5% deoxycholate are present (Fig. 2C, right panel). Second, RIPA buffer does not affect the ability of CFTR IgG (anti-NBD-R) to immunoprecipitate CFTR compared with PBS, 0.2% Triton X-100 buffer (data not shown). Therefore, after cross-linking and lysing the cells in RIPA buffer, the only complex that can be immunoprecipitated using CFTR IgG is CFTR or CFTR-interacting partners that are cross-linked into a higher order complex. Therefore, we cross-linked Calu-3 cells with 0.5 or 1 mM DSP and lysed the cells with RIPA buffer. The protein lysates were immunoprecipitated using NBD-R polyclonal IgG. As shown in Fig. 2D, NHERF1 was detected in the complex with CFTR only when cross-linked into a complex (with DSP). We also detected NHERF2 in the complex, although the signal was much weaker (data not shown). Apical colocalization of CFTR with NHERF1 was also demonstrated biochemically in polarized Calu-3 cells (see "Experimental Procedures" for details). Cross-linked (1 mM DSP) and apically biotinylated membranes were captured using streptavidin-agarose. These membranes were immunoprecipitated using CFTR antibody (anti-NBD-R IgG). As shown in Fig. 2E, CFTR interacted with NHERF1 in the apical membrane. Non-immune IgG was used as negative control.



View larger version (54K):
[in this window]
[in a new window]
  		FIG. 2.
CFTR can exist as a higher order complex in plasma membrane. A, Calu-3 cells were cross-linked as described under "Experimental Procedures." The cells were lysed in RIPA buffer and blotted for CFTR (R1104). B, Calu-3 cells cross-linked with DSP were lysed in RIPA buffer. 50 µg of total proteins were solubilized with sample buffer in the absence (left) or presence (right) of 2.5% {beta}ME before immunoblotting for CFTR (R1104). C, GST-NHERF1 binding to the CFTR in PBS, 0.2% Triton X-100 buffer or RIPA buffer. BHK-CFTR cells were lysed with PBS, 0.2% Triton X-100 (TX100) or RIPA buffer, subjected to pull-down using GST-NHERF1, and immunoblotted with anti-CFTR antibody (R1104). D, BHK-CFTR cells overexpressing NHERF1 were cross-linked with different concentrations of DSP. The cells were lysed in RIPA buffer and co-immunoprecipitated using anti-NBD-R ({alpha}-NBD-R) IgG as described before (19). The immunoprecipitated proteins were then subjected to SDS-PAGE, and the membrane was cut and blotted for CFTR (R1104, upper panel) and NHERF1 (bottom panel), respectively. E, colocalization of CFTR and NHERF1 in apical membranes of polarized Calu-3 cells (see "Experimental Procedures" for details). F, stoichiometry of CFTR:NHERF1 in the complex. Calu-3 cells were cross-linked with 1 mM DSP and subjected to co-immunoprecipitation using anti-NBD-R IgG. The immunoprecipitated complex was treated with 2.5% {beta}ME to dissociate the cross-linked complex and resolved on 10% SDS-PAGE. Purified CFTRHis10 was used as a standard to quantitate CFTR (upper panel), and GST-PDZ2 of NHERF1 was used as a standard to quantitate the co-immunoprecipitated NHERF1 (bottom panel) (20). G, stoichiometry of CFTR:NHERF1 in the purified plasma membrane from Fig. 5C. The plasma membrane was cross-linked with 1 mM DSP and subjected to immunoblotting as in F. H, stoichiometry of CFTR:NHERF1 in the biotinylated apical plasma membrane of polarized Calu-3 cells (see "Experimental Procedures" for details). DST, disuccinimidyl tartarate; IP, immunoprecipitation.

 
Since CFTR polypeptides could form a multimeric complex in the plasma membrane, we next determined the stoichiometry of CFTR:NHERF1 in the complex. For these experiments, we cross-linked Calu-3 cells with 1 mM DSP, lysed the cells using RIPA buffer, and subjected them to immunoblotting (Fig. 2F). Purified CFTRHis10 was used as a standard to quantitate CFTR in the complex, and GST-PDZ2 (of NHERF1) was used as a standard to quantitate NHERF1 (Fig. 2F). We estimated the amount of CFTR and NHERF1 from the standard curve by densitometric analysis using the software Quantity One (Bio-Rad). As shown in Fig. 2F, the CFTR to NHERF1 stoichiometry is 0.058 to 0.001 pmol. We also quantitated the CFTR and NHERF1 amounts present in the purified plasma membrane of Calu-3 cells (also refer to Fig. 5C). The plasma membrane was cross-linked with 1 mM DSP and subjected to immunoblotting as in Fig. 2F. The CFTR to NHERF1 stoichiometry is 0.06 to 0.0015 pmol (Fig. 2G). The relative amount of CFTR and NHERF1 in the biotinylated apical plasma membrane from Fig. 2E also shows a similar stoichiometry (0.0699 to 0.0014 pmol, Fig. 2H). Cumulatively these results provide clear evidence that PDZ domain-containing partner NHERF1 is not the major component of the macromolecular complex in native cell membrane and contributes ~2% of the total cross-linked complex. NHERF2 was also quantitated in this manner, and it contributes to even less than 0.25% of the complex (data not shown). We therefore conclude that (a) in cells expressing CFTR the protein exists as a higher order complex and (b) the PDZ domain proteins (NHERFs) are not the major components of the complex (~2%).



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 5.
Concentration and time dependence of cross-linking. Concentration dependence (A) and time dependence (B) of cross-linking by DSP in Calu-3 cells. Cells were cross-linked, lysed with RIPA buffer, and subjected to Western blotting using R1104 IgG. C, expression of CFTR in various subcellular fractions. An equal amount of protein was loaded in each lane. The blot was probed with anti-CFTR mAb (R1104). D, plasma membrane from C was cross-linked with DSP for 5 min as in A and demonstrated concentration dependence of DSP as well. E, iodide efflux in Calu-3 monolayers (n = 3) subjected to cross-linking with DSP. DSP was added to the nitrate solution (arrow 1), cells were washed four times with the 136 mM nitrate solution containing the cross-linker, and the fractions were collected. The agonist (200 µM cpt-cAMP, 10 µM forskolin, and 100 µM 3-isobutyl-1-methylxanthine) was added to the nitrate solution (with or without DSP, arrow 2). The cells were lysed at the end of the experiment (arrow 3) and subjected to Western blotting using R1104 IgG. F, CFTR function is not affected by low concentration (0-0.5 mM) of DSP and is partially inhibited at 1 mM DSP (upper panel). Data are presented as the percentage of maximal cpt-cAMP stimulation (n = 4). After the assay (E, arrow 3), the cells were lysed and subjected to Western blot analysis using anti-CFTR antibody R1104 (lower panel). G, activation of Cl- secretion by adding cAMP mixture to the apical surface of polarized Calu-3 cells mounted in an Ussing chamber pretreated with or without DSP as described under "Experimental Procedures." The short circuit current was attenuated by 500 mM DPC added to the apical side. H, activation of cAMP mixture-mediated Cl- currents in HT29-CL19A monolayers pretreated with or without DSP as in G.

 
Cross-linked Complex of CFTR in the Plasma Membrane Is Likely to Be Dimers—We further characterized the cross-linked complex. We determined the molecular weight of the CFTR multimer by applying the principle that electrophoretic mobility of proteins in SDS-PAGE is proportional to the logarithm of their mass. We measured the distance from the top of the membrane to the middle of the upper (U) and lower (L) bands (Fig. 3A). Using inositol 1,4,5-trisphosphate receptor and BIR repeat containing ubiquitin-conjugating enzyme (300 and 530 kDa, respectively) and other prestained markers (10-250 kDa, Bio-Rad) as molecular mass standards, we determined the molecular weight of the cross-linked (U) complex (Fig. 3A) and non-cross-linked (L, Fig. 3A) CFTR. The molecular mass corresponding to the two bands of CFTR was estimated as 330 kDa (U) and 160 kDa (L), respectively (Fig. 3B). The molecular weight of the cross-linked band (U) approximately doubles that of the non cross-linked band (L), suggesting that it is likely to be a dimer. Next using DSP we cross-linked BHK cells expressing CFTRHis10. The cells were lysed in RIPA buffer, and His-tagged CFTR was purified using Talon beads as described under "Experimental Procedures." Just as in Fig. 2B, the purified protein could be cross-linked into a dimer as well, and the dimerized complex was reduced to monomeric form using 2.5% {beta}ME (Fig. 3C). To further demonstrate the dimeric state of this Cl- channel, BHK-CFTRHis10 cells and BHK parental cells were transiently transfected with HA-tagged CFTR. Cells were cross-linked with DSP and then lysed in RIPA buffer. His-tagged CFTR was purified using Talon beads and subjected to immunoblotting using anti-HA IgG. The results suggest that CFTRHis10 can interact with HA-CFTR and exists as a dimer (Fig. 3D). These studies clearly demonstrate that a vast majority of CFTR in cells exist as dimers. However, our experiments do not establish directly the existence of dimeric CFTR in the cell surface. To address this, Calu-3 cells were first cross-linked using 1 mM DSP. Next the cells were surface-biotinylated, lysed in RIPA buffer, and subjected to immunoprecipitation using anti-NBD-R IgG. The blot was first probed using streptavidin-HRP (Fig. 3E, left panel) and demonstrated that surface CFTR also exists as a dimer. The same blot was then probed using CFTR antibody R1104 (Fig. 3E, right panel) to confirm that the biotinylated bands were indeed CFTR. It is abundantly clear from these results that CFTR exists as a dimer in the surface plasma membrane as well.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 3.
Cross-linked complex of CFTR in the plasma membrane is likely to be dimers. A, Calu-3 cells were cross-linked with 0.5 mM DSP. 50 µg of protein lysates were blotted using anti-CFTR IgG (R1104). L represents lower band or non-cross-linked CFTR; U represents upper band or cross-linked CFTR. The representative blots of inositol 1,4,5-trisphosphate receptor (IP3R) (300 kDa) and BIR repeat containing ubiquitin-conjugating enzyme (530 kDa) are shown in the middle and right panels, respectively. B, the molecular mass of CFTR was estimated as 330 kDa (U) and 160 kDa (L), respectively, by plotting the mobility against logarithm of molecular weight of different markers (log(molecular weight)). C, BHK cells expressing CFTRHis10 were cross-linked with 1 mM DSP and lysed in RIPA buffer, and His-tagged CFTR was purified using Talon beads as described under "Experimental Procedures." The protein was eluted from the beads using 100 mM imidazole buffer (pH 7.0) containing 0.2% Triton X-100 and blotted with R1104 after treatment with or without {beta}ME. D, BHK-CFTRHis10 cells and BHK parental cells were transiently transfected with HA-tagged CFTR. Cells were then cross-linked with 1 mM DSP and lysed in RIPA buffer. His-tagged CFTR was purified using Talon beads as described under "Experimental Procedures." The eluted protein was subjected to immunoblotting using anti-HA IgG. E, Calu-3 cells were cross-linked with 1 mM DSP, surface-biotinylated, and subjected to immunoprecipitation and immunoblotting as described under "Experimental Procedures." The blot was first probed using streptavidin-HRP (left panel). The same blot was then probed using CFTR antibody R1104 (right panel). {alpha}-NBD-R, anti-NBD-R; IP, immunoprecipitation.

 
CFTR Dimers Dissociate into Monomers upon Solubilization of Membrane with Detergents—Using sucrose gradient sedimentation, Riordan's group (10) elegantly demonstrated that CFTR indeed exists in a monomeric state in detergent-solubilized extract. Using this very same method, we subjected lysates of Calu-3 cell (pretreated with or without cross-linkers) to velocity sedimentation in the linear, continuous sucrose gradients ranging from 10 to 36%. Fig. 4A shows the major band migrating at fraction 8 (upper panel). In this particular experiment, cells were lysed in RIPA buffer. We also lysed the cells with non-ionic detergents such as Triton X-100 or Nonidet P-40 and observed that the band migrated at fraction 8 (data not shown). Recombinant CFTR from BHK-CFTR and BHKCFTRHis10 cells also demonstrated a sedimentation pattern nearly identical to that in Calu-3 cells under non-cross-linked condition (data not shown). The molecular mass was estimated to be ~145 kDa (Fig. 4B). Our results (Fig. 4, A, upper panel, and B) are consistent with those of Riordan's group (10) where they estimated the molecular mass of the detergent-solubilized CFTR to be in the range of 130-200 kDa when using non-ionic detergents (0.09% Nonidet P-40 or 0.2% Triton X-100). However, upon cross-linking Calu-3 cells with either 1 mM DSP or DFDNB and then subjecting them to sucrose velocity gradient as described above, we were able to demonstrate that the major CFTR band migrated at fraction 11 (Fig. 4A, middle and lower panels). The molecular mass of the cross-linked CFTR complex (fraction 11) is ~300 kDa (Fig. 4B), which is twice that of the non-cross-linked complex. These results are consistent with those determined by mobility shift method after cross-linking (Fig. 3, A and B). Fig. 4C shows that a 10-36% linear, continuous sucrose gradient was achieved in these experiments (fraction number versus sucrose concentration). Our studies clearly demonstrate that CFTR is a dimer in native cells and is dissociated into monomers upon solubilization by detergents (SDS, Triton X-100, Nonidet P-40, etc.), which we believe is the reason that Riordan's group failed to detect dimeric CFTR in native cells. More recently, Bear's group (15) has been able to demonstrate that CFTR could exist as dimers in the baculovirus expression system using size exclusion chromatography and pentafluoro-octanoic acid gels.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 4.
CFTR dimers dissociate into monomers upon solubilization of membrane with detergents. A, partially purified membrane of CFTR was solubilized in RIPA buffer and subjected to velocity sedimentation in the linear, continuous sucrose gradients ranging from 10 to 36% as described under "Experimental Procedures," and fractions were blotted using R1104 IgG. A shift of CFTR-containing from fractions 8 of control (upper panel, without cross-linker) to 11 upon cross-linking cells with 1 mM DFDNB (middle panel) or 1 mM DSP (bottom panel) was observed. B, standard curve (r = 0.90) derived from sucrose sedimentation of molecular mass standards (Bio-Rad) in 10-36% linear, continuous sucrose gradient. The standard components were: thyroglobulin, 670 kDa; bovine {gamma} globulin, 158 kDa; chicken ovalbumin, 44 kDa; equine myoglobin, 17 kDa. C,a 10-36% linear, continuous sucrose gradient was observed using a handheld 0-10° Brix refractometer (Fisher) after a 1:3 sample dilution.

 
CFTR Dimerization in Plasma Membrane Is Regulated by Phosphorylation of the Channel—To further characterize the cross-linked CFTR complex, we treated Calu-3 cells with varying concentrations of DSP for varying lengths of time. With the increase in concentration (Fig. 5A) or incubation time of cross-linker (Fig. 5B), CFTR dimer formation increased linearly until almost all of it appeared as the higher molecular weight complex at 2.5 mM DSP. Similar results were observed using DFDNB as cross-linker (data not shown). Next we investigated the CFTR expression in purified plasma membrane and other subcellular fractions (Fig. 5C). The majority of the CFTR polypeptides are present in the plasma membrane and 40,000 x g pellet as reported earlier (32, 33). We cross-linked the plasma membrane fraction with varying concentrations of DSP for 5 min and found kinetics of dimer formation (Fig. 5D) similar to that observed in native Calu-3 cells (Fig. 5A).

CFTR Cl- channel function was monitored using iodide efflux assay, and the effect of DSP on the channel function in Calu-3 cells was studied (Fig. 5E). The cells were first loaded with 136 mM NaI for 1 h and then washed five times with 136 mM NaNO3. DSP was added to the nitrate solution (Fig. 5E, arrow 1), cells were washed four more times with the 136 mM nitrate solution containing the cross-linker, and the fractions were collected. The agonist cAMP mixture (200 µM cpt-cAMP, 10 µM forskolin, and 100 µM 3-isobutyl-1-methylxanthine) was added to the nitrate solution containing DSP (Fig. 5E, arrow 2). The cells were lysed at the end of the experiment (Fig. 5E, arrow 3) and subjected to Western blotting. Fig. 5F shows the functional activity of CFTR Cl- channel (n = 3) using a wide range concentrations of DSP. The channel activity was not affected at lower concentrations (0-0.5 mM DSP) and was partially inhibited at 1 mM DSP (Fig. 5F, upper panel). After the assay (Fig. 5E, arrow 3), the cells were lysed and subjected to Western blotting analysis using R1104 antibody (Fig. 5F, lower panel). As can be seen in Fig. 5F, the active form of the channel exists as a dimer. A functional study using polarized epithelial cells mounted in an Ussing chamber was also performed. Calu-3 cells and HT29-CL19A cells were cross-linked with 0.5 mM DSP for 10 min and then mounted in an Ussing chamber. CFTR channel was activated using the cAMP mixture. As shown in Fig. 5, G and H, the short circuit current in response to these reagents did not show a significant difference between the control and DSP-pretreated group. DPC was used to inhibit the CFTR Cl- channel. It is likely that the active form of the channel exists as a dimer as well.

It was therefore of interest to see whether phosphorylation of the channel had an effect on the dimerization of CFTR. Toward this goal, cells were pretreated with varying concentrations of cpt-cAMP (cell membrane-permeable analog of cAMP) for 2 min and then cross-linked with 0.5 mM DSP for 10 min. Interestingly, in cultured Calu-3 cells, the dimer formation of CFTR polypeptides was increased with the increasing concentration of cpt-cAMP (Fig. 6A). The cAMP mixture had the maximum effect on CFTR dimerization. The channels were phosphorylated under this condition (data not shown). We next asked whether the increased dimerization was due to phosphorylation or the functionally active form of the channel. To this end, we first treated the cells with cAMP-dependent protein kinase inhibitor (H89). Our results demonstrated that the increased dimerization observed in the presence of cpt-cAMP can be inhibited by H89, suggesting that phosphorylation by cAMP-dependent protein kinase is essential for increased dimer formation (Fig. 6B). In parallel, the functional activity of CFTR Cl- channel was monitored in the presence of H89 and showed that the channel was indeed inhibited (Fig. 6C). Next we investigated the effect of CFTR Cl- channel inhibitor (DPC) on dimerization of CFTR polypeptides in the presence of the cpt-cAMP. Fig. 6D shows that there is almost no significant difference between the dimerization patterns of DPC-treated cells and control cells upon activation with cpt-cAMP. The iodide efflux shows that DPC can inhibit the activation of CFTR Cl- channel by cpt-cAMP (Fig. 6E). These experiments clearly demonstrated that dimerization of CFTR polypeptides in cultured cells is cAMP-dependent protein kinase-dependent and DPC-insensitive, which implies that phosphorylation of the Cl- channel, rather than its active form, is crucial for dimer formation.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 6.
CFTR dimerization in cells is regulated by phosphorylation but not by channel activity. A, Calu-3 cells were pretreated with varying concentrations of cpt-cAMP for 2 min followed by incubation with 0.5 mM DSP for 5 min. The cells were lysed in RIPA buffer and subjected to Western blot using R1104 IgG. B, Calu-3 cells were treated with 1 µM H89 (cAMP-dependent protein kinase inhibitor) for 2 min, then cpt-cAMP for 2 min (middle), and cpt-cAMP alone for 2 min (right) followed by incubation with 0.5 mM DSP. The cells were lysed in RIPA buffer and subjected to Western blot using R1104 IgG. The upper panel is a representative blot of four individual experiments. The bar graph (n = 4) is also shown (bottom panel). C, iodide efflux in the absence or presence of H89. The data are presented as the percentage of maximal cpt-cAMP stimulation (n = 4). D, Calu-3 cells were pretreated with 500 µM DPC (CFTR Cl- channel inhibitor) for 2 min, then cpt-cAMP for 2 min (middle), and cpt-cAMP alone for 2 min (right) followed by incubation with 0.5 mM DSP. The cells were lysed in RIPA buffer and subjected to Western blot using R1104 IgG. The upper panel is a representative blot of three individual experiments. The bar graph (n = 3) is also shown (bottom panel). E, iodide efflux measured in the absence or presence of DPC. The data are presented as the percentage of maximal cpt-cAMP stimulation (n = 3).

 
PDZ Motif-based Interaction Is Not Essential for Dimerization of CFTR Polypeptides—To investigate whether dimerization of CFTR proteins requires the C-tail PDZ motif, we treated BHK-CFTR and BHK-CFTRHis10 cells with cross-linkers. We found no significant difference between the dimerization kinetics of BHK-CFTR cells and BHK-CFTRHis10 cells treated with DSP at various concentrations (Fig. 7A), indicating that the CFTR dimerization is C-tail PDZ motif-independent, namely NHERF1-independent. We demonstrated that the CFTRHis10 cannot interact with NHERF2 using a pull-down assay (Fig. 7B). COS-7 cells were transiently transfected with pcDNA3-WT-CFTR and pcDNA3-{Delta}TRL-CFTR using LipofectAMINE 2000 (Invitrogen) as described before (5). Deletion of the PDZ binding motif of CFTR (last 3 amino acids, referred to as {Delta}TRL-CFTR) eliminated the physical interaction between CFTR and NHERF1 as demonstrated by pull-down assay (Fig. 7C), suggesting that the PDZ motif of CFTR is essential for the physical interaction between CFTR and NHERF1. Cross-linking patterns between WT-CFTR and {Delta}TRL-CFTR were very similar and did not show any significant difference in response to 0.5 and 1.0 mM DSP at 37 °C for 10 min (Fig. 7D). Both BHK-CFTR and BHK-CFTRHis10 cells transiently transfected with NHERF1 (or NHERF2, data not shown) did not show any significant differences in the dimerization kinetics. A representative blot is shown in Fig. 7E where BHK-CFTR cells with or without overexpressing NHERF1 were cross-linked with varying concentrations of DSP and showed the similar pattern in cross-linking complex. The NHERF1 expression is shown in Fig. 7E, bottom panel. We also co-immunoprecipitated lysates from BHK-CFTR and BHK-CFTRHis10 cells both transiently transfected with NHERF1, subjected them to Western blotting, and found that the protein interactions were indeed occurring in the BHK-CFTR cells (Fig. 7F, left bottom panel) but not in the BHK-CFTRHis10 cells (Fig. 7F, right bottom panel) after cross-linking with 0.5 mM DSP. These results, together with the dimerization of purified CFTRHis10 as shown in Fig. 3C, clearly demonstrate that although NHERF1 and -2 can be in a complex with CFTR, they are not likely to contribute to the dimerization of the channel. G{beta} and Syntaxin 1A Are Also Parts of the Multimer Complex of CFTR Polypeptides—Given that CFTR exists mostly as a dimer and a very small portion of this complex is NHERF (<2%), it is of interest to identify functionally relevant proteins that could be associated with this complex. Previously we demonstrated that CFTR can be in a complex with {beta}2 adrenergic receptor via NHERF1-mediated interaction (5). Since {beta}2 adrenergic receptor can be in the complex, some obvious candidates include the G-protein signaling components. We therefore over-expressed {beta}2 adrenergic receptor in COS-7 cells; lysed the cells in PBS, 0.2% Triton X-100; and subjected them to pull-down using GST or GST-NHERF1 (1 µM). Bound proteins were subjected to Western blot analysis using affinity-purified G{beta} IgG. As shown in Fig. 8A, right panel, the amount of G{beta} subunit associated with GST-NHERF1 increased following overexpression of {beta}2 adrenergic receptor, suggesting a physical association with the complex presumably mediated through {beta}2 adrenergic receptor. The inputs are shown in Fig. 8A, left panel. By performing two-dimensional electrophoresis of the cross-linked CFTR immunoprecipitate from Calu-3 cells, we further confirmed that G{beta} can be in the complex of CFTR (Fig. 8B, lower panel). We detected an immunoreactive band of approximately pI 5.6 and molecular mass of 37 kDa. Non-immune IgG was used as control and is shown in Fig. 8B, upper panel. We demonstrated previously that syntaxin 1A, a SNARE protein essential for the docking and fusion of exocytotic vesicles, binds to the N-terminal tail of CFTR and down-regulates CFTR function (34). It was therefore of interest to determine whether this SNARE protein could be associated with the CFTR multimer complex. Syntaxin 1A {Delta}C (deletion of the C-terminal membrane anchor) bound directly to CFTR and CFTRHis10 with similar affinities (Fig. 8C), suggesting that the N-terminal tail of CFTR is essential for the protein-protein interaction and not the C-tail (18). Syntaxin 1A was transiently expressed in BHKCFTR or BHK-CFTRHis10 cells. The cells were lysed and subjected to pull-down using GST-NHERF1 as bait. As shown in Fig. 8D, GST-NHERF1 could pull down CFTR and syntaxin 1A simultaneously if the PDZ motif was intact, suggesting that syntaxin 1A can interact with CFTR and indirectly with GSTNHERF1. Syntaxin 1A, however, was not pulled down by GSTNHERF1 when associated with CFTRHis10. The inputs are shown in Fig. 8D, right panel. Although the opposing tails of CFTR bind distinct proteins, our studies demonstrate that these proteins can co-exist in a macromolecular complex. Using two-dimensional gel electrophoresis, we further showed syntaxin 1A (~pI 5.1, ~33 kDa) can be co-immunoprecipitated with CFTR in a cross-linked complex (Fig. 8E). We have not quantitated the relative stoichiometry of G{beta} and syntaxin 1A in the CFTR multimer complex.



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 7.
PDZ motif-based interaction is not essential for dimerization of CFTR polypeptides. A, BHK-CFTR and BHK-CFTRHis10 cells were cross-linked with DSP, lysed in RIPA buffer, resolved on a 5% gel, and subjected to immunoblotting for CFTR using R1104 IgG (upper panel). The upper panel is a representative blot of three individual experiments. The bar graph (n = 3) is also shown (bottom panel). B, BHK-CFTR and BHK-CFTRHis10 cells were lysed with PBS, 0.2% Triton X-100, subjected to pull-down using GST-NHERF2, and probed for CFTR. C, COS-7 cells were transiently transfected with pcDNA3-WT-CFTR and pcDNA3-{Delta}TRL-CFTR using LipofectAMINE 2000 as described before (5). The cells were lysed with PBS, 0.2% Triton X-100, subjected to pull-down using GST-NHERF1, and probed for CFTR. D, COS-7 cells expressing WT-CFTR or {Delta}TRL-CFTR were cross-linked with 0.5 or 1.0 mM DSP followed by immunoblotting as in A. E, BHK-CFTR cells and BHK-CFTR cells overexpressing NHERF1 were cross-linked with varying concentrations of DSP, lysed in RIPA buffer, resolved on a 4-15% gel, and subjected to immunoblotting for CFTR using R1104 IgG (upper panel). The NHERF1 expression is shown in the bottom panel. F, cell lysates from BHK-CFTR and BHK-CFTRHis10 cells both transiently transfected with NHERF1 with or without cross-linker (0.5 mM DSP) were co-immunoprecipitated with anti-NBD-R IgG and blotted using either CFTR mAb (R1104) or NHERF1 mAb.

 



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 8.
G{beta} and syntaxin 1A are also parts of the multimer complex of CFTR polypeptides. A, COS-7 cells overexpressing {beta}2 adrenergic receptor were lysed with PBS, 0.2% Triton X-100 and subjected to pull-down using GST or GST-NHERF1 (1 µM). The complex was subjected to Western blot analysis and probed using affinity-purified anti-G{beta} IgG (1:500). B, Calu-3 cells were cross-linked with 1 mM DSP, lysed in RIPA buffer, and subjected to co-immunoprecipitation using mouse anti-CFTR antibody (MM13-4, bottom panel) or normal mouse IgG (upper panel). The samples were eluted and subjected to two-dimensional gel electrophoresis according to the manufacturer's instructions (Bio-Rad) as described under "Experimental Procedures." Isoelectric focusing (pI 3-10) was performed in the first dimension, SDS-PAGE was performed in a 10% polyacrylamide gel in the second dimension, and proteins were subjected to Western blotting with affinity-purified rabbit G{beta} IgG. Standard markers for two-dimensional gel electrophoresis (Pierce) were used in parallel (data not shown). C, syntaxin 1A {Delta}C binds directly to CFTR and CFTRHis10 with similar affinity. BHK-CFTR or BHK-CFTRHis10 cells were lysed with PBS, 0.2% Triton X-100, subjected to pull-down using GST or GST-syntaxin 1A {Delta}C (0.10-1.00 µM), and immunoblotted with monoclonal CFTR antibody (R1104). D, CFTR can interact with NHERF1 and syntaxin 1A simultaneously. BHK-CFTR or BHK-CFTRHis10 cells were transfected with syntaxin 1A (vaccinia virus expression system, Ref. 18). The cells were lysed, subjected to pull-down using GST or GST-NHERF1 (0.14-0.57 µM), and probed for syntaxin 1A (14D8 monoclonal antibody, Ref. 20) and monoclonal CFTR antibody (R1104). E, two-dimensional gel electrophoresis similar to that in B (bottom panel) except that affinity-purified anti-syntaxin 1A IgG (14D8) was used for blotting. Syn, syntaxin.

 
In summary, based on our studies and published data, we hypothesize the following. (i) CFTR in epithelial cells can be clustered to distinct microdomains (in apical membrane) by PDZ-based interactions as reported for other channels (3, 35). (ii) CFTR polypeptides may form multimeric complexes in such microdomains, and a large portion of CFTR is likely to exist as PDZ protein-independent dimers. (iii) Various PDZ domain-interacting partners such as other channels (e.g. epithelial Na+ channel, Ref. 36), receptors (e.g. {beta}2 adrenergic receptor, Ref. 5), and cytoskeleton proteins (e.g. ezrin, Ref. 37) can be in the complex as well. (iv) Other related proteins that do not interact directly with the PDZ motif but rather through other intermediary proteins (e.g. G{beta} and syntaxin 1A) can be in the complex as well and are likely to be involved in signaling (18, 20, 34). (v) Upon signaling, a local increase in cAMP leads to the phosphorylation of the CFTR Cl- channel, resulting in rapid dimerization of the channel, which probably translocates the dimers away from the complex leading to the Cl- efflux from the channel (5).


   FOOTNOTES
 
* This work was supported by National Institutes of Health Grant DK58545 (to A. P. N.) and American Heart Association Southeast Affiliate Grant-in-aid 0265008B (to A. P. N.). 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. Back

{ddagger} To whom correspondence should be addressed: Dept. of Physiology, University of Tennessee Health Science Center, 420 NASH, 894 Union Ave., Memphis, TN 38163. Tel.: 901-448-3137; Fax: 901-448-7126; E-mail: anaren{at}utmem.edu.

1 The abbreviations used are: PDZ, PSD-95/Dlg/ZO-1; CFTR, cystic fibrosis transmembrane conductance regulator; MBP, maltose-binding protein; HA, hemagglutinin; NHERF, Na+/H+ exchanger regulatory factor; BHK, baby hamster kidney; cpt-cAMP, 8-(4-chlorophenylthio)-cAMP; DSP, dithiobis(succinimidyl propionate); DFDNB, 1,5-difluoro-2,4-dinitrobenzene; {beta}ME, {beta}-mercaptoethanol; DPC, diphenylamine-2-carboxylate; GST, glutathione S-transferase; CHAPSO, 3-[(3-cholamidopropyl)dimethylammonio]-2-hydroxy-1-propanesulfonic acid; mAb, monoclonal antibody; PBS, phosphate-buffered saline; RIPA, radioimmune precipitation assay; HRP, horseradish peroxidase; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; U, upper; L, lower; WT, wild type; SNARE, soluble N-ethylmaleimide-sensitive factor attachment protein receptor; sulfo-NHS-LC-biotin, succinimidyl-6-(biotinamido)hexanoate; biotin-XX-hydrazide, 6-((6-((bioti-noyl)amino)hexanoyl)amino)hexanoic acid, hydrazide; EGS, ethylene glycol bis(succinimidyl succinate); DTBP, dimethyl 3,3'-dithiobis(propionimidate); DTSSP, 3,3'-dithiobis(sulfosuccinimidyl propionate). Back


   ACKNOWLEDGMENTS
 
We thank Dr. David L. Armbruster (Scientific Publications, University of Tennessee Health Science Center) for critically reading the manuscript and Dr. Randy Hall (Emory University) for kindly providing NHERF2 cNDA.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 REFERENCES
 

  1. Fanning, A. S., and Anderson, J. M. (1999) Curr. Opin. Cell Biol. 11, 432-439[CrossRef][Medline] [Order article via Infotrieve]
  2. Riordan, J. R., Rommens, J. M., Kerem, B., Alon, N., Rozmahel, R., Grzelczak, Z., Zielenski, J., Lok, S., Plavsic, N., Chou, J. L., Drumm, M., Iannuzi, M., Collins, F. S., and Tsui, L-C. (1989) Science 245, 1066-1073[Abstract/Free Full Text]
  3. Huang, P., Lazarowski, E. R., Tarran, R., Milgram, S. L., Boucher, R. C., and Stutts, M. J. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 14120-14125[Abstract/Free Full Text]
  4. Weinman, E. J., Steplock, D., Wang, Y., and Shenolikar, S. (1995) J. Clin. Investig. 95, 2143-2149
  5. Naren, A. P., Cobb, B., Li, C., Roy, K., Nelson, D., Heda, G. D., Liao, J., Kirk, K. L., Sorscher, E. J., Hanrahan, J., and Clancy, J. P. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 342-346[Abstract/Free Full Text]
  6. Zhu, T., Dahan, D., Evagelidis, A., Zheng, S.-X., Luo, J., and Hanrahan, J. W. (1999) J. Biol. Chem. 274, 29102-29107[Abstract/Free Full Text]
  7. Hallows, K. R., Raghuram, V., Kemp, B. E., Witters, L. A., and Foskett, J. K. (2000) J. Clin. Investig. 105, 1711-1721[Medline] [Order article via Infotrieve]
  8. Hallows, K. R., Kobinger, G. P., Wilson, J. M., Witters, L. A., and Foskett, J. K. (2003) Am. J. Physiol. 284, C1297-C1308
  9. Marshall, J., Fang, S., Ostedgaard, L. S., O'Riordan, C. R., Ferrara, D., Amara, J. F., Hoppe, H., IV, Scheule, R. K., Welsh, M. J., and Smith, A. E. (1994) J. Biol. Chem. 269, 2987-2995[Abstract/Free Full Text]
  10. Chen, J. H., Chang, X. B., Aleksandrov, A. A., and Riordan, J. R. (2002) J. Membr. Biol. 188, 55-71[CrossRef][Medline] [Order article via Infotrieve]
  11. Eskandari, S., Wright, E. M., Kreman, M., Starace, D. M., and Zampighi, G. A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 11235-11240[Abstract/Free Full Text]
  12. Zerhusen, B., Zhao, J., Xie, J., Davis, P. B., and Ma, J. (1999) J. Biol. Chem. 274, 7627-7630[Abstract/Free Full Text]
  13. Raghuram, V., Mak, D.-O. D., and Foskett, J. K. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 1300-1305[Abstract/Free Full Text]
  14. Ramjeesingh, M., Li, C., Kogan, I., Wang, Y., Huan, L.-J., and Bear, C. E. (2001) Biochemistry 40, 10700-10706[CrossRef][Medline] [Order article via Infotrieve]
  15. Ramjeesingh, M., Kidd, J. F., Huan, L. J., Wang, Y., and Bear, C. E. (2003) Biochem. J. 374, 793-797[CrossRef][Medline] [Order article via Infotrieve]
  16. Hall, R. A., Ostedgaard, L. S., Premont, R. T., Blitzer, J. T., Rahman, N., Welsh, M. J., and Lefkowitz, R. J. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 8496-8501[Abstract/Free Full Text]
  17. Tabcharani, J. A., Low, W., Elie, D., and Hanrahan, J. W. (1990) FEBS Lett. 270, 157-164[CrossRef][Medline] [Order article via Infotrieve]
  18. Naren, A. P., Quick, M. W., Collawn, J. F., Nelson, D. J., and Kirk, K. L. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 10972-10977[Abstract/Free Full Text]
  19. Naren, A. P. (2002) Methods Mol. Med. 70, 175-186[Medline] [Order article via Infotrieve]
  20. Naren, A. P., Di, A., Cormet-Boyaka, E., Boyaka, P. N., McGhee, J. R., Zhou, W., Akagawa, K., Fujiwara, T., Thome, U., Engelhardt, J. F., Nelson, D. J., and Kirk, K. L. (2000) J. Clin. Investig. 105, 377-386[Medline] [Order article via Infotrieve]
  21. DeMarco, S. J., Chicka, M. C., and Strehler, E. E. (2002) J. Biol. Chem. 277, 10506-10511[Abstract/Free Full Text]
  22. Prince, L. S., Workman, R. B., Jr., and Marchase, R. B. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 5192-5196[Abstract/Free Full Text]
  23. Inoue, G., Cheatham, B., and Kahn, C. R. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 14919-14924[Abstract/Free Full Text]
  24. Smith, J. J., Karp, P. H., and Welsh, M. J. (1994) J. Clin. Investig. 93, 307-311
  25. Chang, X. B., Tabcharani, J. A., Hou, Y. X., Jensen, T. J., Kartner, N., Alon, N., Hanrahan, J. W., and Riordan, J. R. (1993) J. Biol. Chem. 268, 11304-11311[Abstract/Free Full Text]
  26. Wang, S., Yue, H., Derin, R. B., Guggino, W. B., and Li, M. (2000) Cell 103, 169-179[CrossRef][Medline]