HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhu, T. Articles by Hanrahan, J. W. Search for Related Content PubMed PubMed Citation Articles by Zhu, T. Articles by Hanrahan, J. W. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 41, 29102-29107, October 8, 1999

Association of Cystic Fibrosis Transmembrane Conductance Regulator and Protein Phosphatase 2C^*

T. Zhu^§, D. Dahan^§, A. Evagelidis, S.-X. Zheng, J. Luo^¶, and J. W. Hanrahan^**

From the Department of Physiology, McGill University, Montréal, Québec H3G 1Y6, Canada

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cystic fibrosis transmembrane conductance regulator (CFTR) chloride channels are rapidly deactivated by a membrane-bound phosphatase activity. The efficiency of this regulation suggests CFTR and protein phosphatases may be associated within a regulatory complex. In this paper we test that possibility using co-immunoprecipitation and cross-linking experiments. A monoclonal anti-CFTR antibody co-precipitated type 2C protein phosphatase (PP2C) from baby hamster kidney cells stably expressing CFTR but did not co-precipitate PP1, PP2A, or PP2B. Conversely, a polyclonal anti-PP2C antibody co-precipitated CFTR from baby hamster kidney membrane extracts. Exposing baby hamster kidney cell lysates to dithiobis (sulfosuccinimidyl propionate) caused the cross-linking of histidine-tagged CFTR (CFTR_His10) and PP2C into high molecular weight complexes that were isolated by chromatography on Ni²⁺-nitrilotriacetic acid-agarose. Chemical cross-linking was specific for PP2C, because PP1, PP2A, and PP2B did not co-purify with CFTR_His10 after dithiobis (sulfosuccinimidyl propionate) exposure. These results suggest CFTR and PP2C exist in a stable complex that facilitates regulation of the channel.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CFTR¹ is a tightly regulated chloride channel expressed in epithelial and other cell types (1). CFTR channels become deactivated within ~10 s after membrane patches are excised from cAMP-stimulated Chinese hamster ovary cells into bath solution containing ATP at 37 °C (2). Similar results are obtained using Chinese hamster ovary and airway epithelial cell lines at 20 °C except the deactivation requires ~100 s (3). Channel rundown is caused by dephosphorylation of PKA sites, because it does not occur if the PKA catalytic subunit is present in the bath solution, and deactivated channels can be restimulated by exposure to PKA (2, 4). These results indicate that at least one of the phosphatases regulating CFTR is membrane-delimited.

The phosphatases that control CFTR have not yet been identified definitively at the molecular level. CFTR is regulated by PKA phosphorylation on multiple serine residues (5-7). Serine/threonine protein phosphatases are classified according to their functional properties: Type 1 phosphatases (PP1) dephosphorylate the  subunit of phosphorylase kinase and are inhibited by thermostable protein inhibitors 1 and 2. Type 2 phosphatases (PP2) dephosphorylate the  subunit of phosphorylase kinase and are insensitive to protein inhibitors 1 and 2 (8). Among the type 2 protein phosphatases, PP2A is distinguished by its sensitivity to okadaic acid and calyculin A, PP2B by its requirement for Ca²⁺ and calmodulin and sensitivity to inhibitors such as deltamethrin, and PP2C by its requirement for relatively high (mM) levels of Mg²⁺.

Several phosphatases dephosphorylate full-length CFTR or recombinant R domain protein in vitro (3, 9, 10), but this does not establish their physiological role because all are usually present in intact cells. The specificity of protein phosphatases depends more on proximity than on the sequence of the target phosphoprotein, hence the importance of determining if CFTR is normally associated with a particular phosphatase.

Functional studies implicate PP2C as a CFTR phosphatase in Chinese hamster ovary, baby hamster kidney (BHK), and airway epithelial cells. CFTR rundown is relatively insensitive to okadaic acid and does not require Ca²⁺ or calmodulin (2) but is inhibited by low Mg²⁺ (11). PP2C is the most potent deactivator of CFTR channels among the exogenous phosphatases tested, and like the endogenous (i.e. membrane-associated) CFTR phosphatase in these cell types, exogenous PP2C inhibits the channel by decreasing the rate at which it enters open bursts. None of the purified phosphatases deactivate CFTR channels completely (11), consistent with the notion that CFTR is regulated by multiple phosphatases (12). Indeed, 5-10% of the activity remains even when patches are exposed to both PP2C and PP2A simultaneously, suggesting they may not be the only players (11).

Rapid deactivation of CFTR channels in excised patches has led to speculation that CFTR and its phosphatase(s) may be physically associated within a regulatory complex (2, 3, 13). In this study we test this possibility in unstimulated BHK cells using co-immunoprecipitations and chemical cross-linking. The results indicate that CFTR is closely associated with PP2C but not PP1, PP2A, or PP2B.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Restriction enzymes were from New England Biolabs (Beverly, MA) or Amersham Pharmacia Biotech. Casein and protein G-Sepharose were from Sigma. Ni²⁺-NTA-agarose was from Qiagen (Chatsworth, CA). Affi-prep protein A columns were from Bio-Rad. pCR-II vector was from Invitrogen (San Diego, CA). Recombinant PP1 catalytic subunit and bovine brain PP2B were from Calbiochem. pGEX-2t vector and Sephadex G-50 columns were from Amersham Pharmacia Biotech. Protein inhibitor-II, polyclonal anti-PP1 (catalog no. 06-221), polyclonal anti-human PP2A (catalog no. 06-222), monoclonal anti-bovine PP2B (catalog no. 05-187), and monoclonal anti-rabbit Na⁺/K⁺-ATPase -1 subunit (catalog no. 05-369) were from Upstate Biotechnology (Lake Placid, NY). [-³²P]ATP and anti-GST antibody were from Amersham Pharmacia Biotech. Monoclonal anti-human R domain antibody (catalog no. 1660-01) was from Genzyme Diagnostics (Cambridge, MA). Protease inhibitor mixture tablets were from Roche Molecular Biochemicals. Dithiobis(sulfosuccinimidyl propionate) (DTSSP) and bicinchoninic protein assay reagents were from Pierce. Peroxidase-conjugated secondary antibodies were from Jackson ImmunoResearch (West Grove, PA). Polyclonal anti-human lactate dehydrogenase 5 antibody (catalog no. 5603-0036) was from Anawa (Zürich, Switzerland). Monoclonal anti-human cytochrome oxidase subunit I antibody (catalog no. A-6405) was from Molecular Probes (Eugene, OR). Protein kinase A from bovine heart was from the laboratory of Dr. M. P. Walsh (University of Calgary, Canada; see Ref. 2 for details). Bacterially expressed human PP2C was kindly provided by Drs. C. Chapmwaud and P. T. Cohen (University of Dundee, United Kingdom). Anti-CFTR antibody (M3A7 (14)) was a gift from Drs. N. Kartner (Dept. of Pharmacology, University of Toronto, Canada) and J. R. Riordan (Mayo Clinic Scottsdale, Scottsdale, AZ). Purified turkey gizzard PP2C and PP2A1 were generously provided by Dr. M. D. Pato (University of Saskatchewan, Saskatoon, Canada). BL21 cells transformed with Paramecium PP2C (15) and detailed instructions regarding its purification were kindly provided by Dr. J. E. Schultz (Tübingen, Germany).

Cells-- BHK and Chinese hamster ovary cells stably transfected with pNUT vector alone or with pNUT vector containing wild-type CFTR (BHK() and BHK-wt, respectively) were provided by Drs. X.-B. Chang and J. R. Riordan (see Refs. 2, 7, and 16). The human airway submucosal gland cell line Calu-3 was obtained from the American Type Culture Collection (Rockville, MD) and was cultured as described previously (17, 18). To confirm PP2C and CFTR expression, cells were lysed by sonication in buffer containing 62.5 mM Tris-HCl, pH 6.8, 0.3% SDS, 10% glycerol, 5% -mercaptoethanol and protease inhibitor mixture and subjected to SDS-PAGE and Western blot analysis.

Cell Fractionation-- Cells were washed with phosphate-buffered saline, harvested by scraping, and centrifuged (1,500 × g, 5 min). Cell pellets were washed with ice-cold phosphate-buffered saline and resuspended in ice-cold lysis buffer (10 mM HEPES, pH 7.2, 2 mM EDTA) supplemented with protease inhibitor mixture. After a 10-min incubation on ice, cells were homogenized with 25 strokes of a tight fitting Dounce homogenizer, and nuclei and mitochondria were removed by centrifugation (4,500 × g, 10 min). Microsomes were collected by centrifugation of the supernatant (100,000 × g, 45 min). The membrane pellet was resuspended in 50 mM HEPES, pH 7.2, and stored at 70 °C.

For phosphorylation and immunoblotting, cells were harvested, washed with cold phosphate-buffered saline, and sonicated for 20 s in twice concentrated (2×) homogenizing buffer (500 mM sucrose, 100 mM Tris-HCl, pH 7.0, 4 mM EDTA, 4 mM EGTA, 2% (w/v) Nonidet P-40, 2 µg/ml phenylmethylsulfonyl fluoride, and protease inhibitor mixture). After centrifugation (7800 × g, 10 min) the supernatant was mixed with an equal volume of glycerol and stored at 20 °C.

Protein Detection-- SDS-PAGE was done according to Ref. 19. Proteins in the gel were detected by staining with Coomassie Blue. Immunoblotting was as described previously (7).

Anti-PP2C Antibodies-- Antibody was raised against a fusion protein consisting of GST fused to amino acids 296-332 of PP2C (GST-PP2C₂₉₆). Reverse transcriptase-polymerase chain reaction was carried out using mRNA from Calu-3 cells as the template. The polymerase chain reaction product was inserted into pCR-II, subcloned into pGEX-2t at the EcoRI site, and confirmed by sequencing. Protease-deficient BL21 cells were transformed with the pGEX-2t plasmids. Fusion protein expression was induced using 0.1 mM isopropyl-1-thio--D-galactopyranoside for 2 h and purified on glutathione-Sepharose 4B as described by the manufacturer. After purification, fusion protein (500 µg in 0.5 ml of phosphate-buffered saline) was emulsified with 0.5 ml of Freund's complete adjuvant and injected subcutaneously into female New Zealand White rabbits. Serum was collected for 6 months and purified on Affi-prep protein A support columns as described by the manufacturer.

Recombinant Paramecium PP2C-- Bacteria (BL21 cells) were transformed with full-length Paramecium PP2C/pET-16b vector (15), and PP2C expression was induced with 1 mM isopropyl-1-thio--D-galactopyranoside for 2 h. Cells were harvested, lysed, and loaded on a Ni²⁺-NTA-agarose column. After extensive washing, PP2C was eluted with buffer containing 1 M imidazole, 500 mM NaCl, 20 mM Tris-HCl, pH 7.9. The yield was 900 µg of PP2C/liter of culture. The phosphatase activity was 17.72 milliunit/mg protein, where 1 unit is the amount of PP2C releasing 1 µmol of ³²PO₄ from phosphocasein/min at 30 °C.

Recombinant CFTR Regulatory (R) Domain-- A 579-nucleotide DNA fragment encoding residues 645-837 was amplified by polymerase chain reaction from pBQ6.2 plasmid DNA (generously provided by Dr. J. M. Rommens, University of Toronto) using Vent polymerase. After ligation into the plasmid pGEX-2t, the sequence was verified, and the GST-R domain fusion protein was expressed and purified as described above for GST-PP2C₂₉₆.

Immunoprecipitations-- For CFTR immunoprecipitations, 3 × 10⁶ cells were washed and resuspended in 1.6 ml of lysis buffer (10 mM HEPES, pH 7.2, 2 mM EDTA, and protease inhibitor mixture). After 10 min on ice the lysate was subjected to 10 strokes in a tight-fitting Dounce homogenizer; 1.6 ml of sucrose buffer (500 mM sucrose, 10 mM HEPES, pH 7.2) was added, and the lysate was homogenized further using 15 strokes. The homogenate was centrifuged (7,800 × g, 10 min) to remove nuclei and unbroken cells, and Triton X-100 was added to 1%.

For PP2C immunoprecipitations, microsomes were prepared as described above under "Cell Fractionation." Membrane proteins were solubilized in 1 ml of solubilization buffer (lysis buffer supplemented with 1% Triton) and centrifuged to remove insoluble material.

Protein concentration of the lysates and membrane fractions was estimated using the Bio-Rad protein assay kit (Richmond, CA). Aliquots of lysate (1 mg) or microsomes (0.4 mg) were diluted in 1 ml of solubilization buffer containing 20 mM magnesium acetate and incubated for 20 min on a rotating shaker with 5 µl of 50% protein G-Sepharose 4B. After centrifugation (7,800 × g, 10 min), supernatants were collected and incubated overnight with 2 µg of anti-PP2C or anti-CFTR antibody. 20 µl of 50% protein G-Sepharose 4B beads were added, and the incubation was extended for 2 h. The beads were washed six times with solubilization buffer containing 20 mM magnesium acetate and resuspended in 100 µl of solubilization buffer. For Western blot analysis, each lane was loaded with 20 µl of the suspension. All steps were carried out at 4 °C.

Expression of CFTR_His10-- Ten histidine codons were added to the 3'-end of CFTR, which was subcloned into pNUT and transfected into BHK cells using calcium phosphate co-precipitation (7). Colonies in which integrated sequences were highly amplified were selected using 500 µM methotrexate, and CFTR_His10 expression was confirmed by immunoblotting. When tested in planar bilayers, channels produced by CFTR_His10 were indistinguishable from those produced by wild-type CFTR (i.e. Cl selective, ~10 picosiemens, voltage-independent, PKA-activated).

DTSSP Cross-linking-- Cells (~1900 cm²) were harvested by scraping, washed with ice-cold phosphate-buffered saline, resuspended in 4 ml of phosphate lysis buffer (50 mM NaH₂PO₄, pH 8.0, 150 mM NaCl, 1% Triton X-100, EDTA-free protease inhibitor mixture), and stirred for 1 h at 4 °C. Insoluble material was removed by centrifugation (15,000 × g, 5 min). Protein concentration of the cell lysate (typically 11-12 mg/ml) was determined by the bicinchoninic assay and adjusted to 4.5 mg/ml with lysis buffer. DTSSP was added to a 5-ml aliquot of lysate from freshly prepared stock (400 µM DTSSP final concentration). After stirring for 1 h at 4 °C, the reaction was stopped using 1 M Tris-HCl, pH 8.0 (10 mM final concentration). Imidazole was added to the sample to a final concentration of 20 mM immediately before purification.

Purification of Cross-linked CFTR_His10 Ni²⁺-- NTA-agarose (0.5 ml) was pre-equilibrated with phosphate lysis buffer containing 20 mM imidazole, incubated batchwise with cross-linked lysate, and stirred for 1 h at 4 °C. The resin was loaded into a column and washed twice with a 4-ml wash buffer containing 40 mM imidazole, 50 mM NaH₂PO₄, pH 8.0, 500 mM NaCl, 0.5% Triton X-100. Proteins were eluted in three 0.5-ml fractions with wash buffer containing 300 mM imidazole. EDTA-free complete protease inhibitor mixture was added to all buffers immediately before use.

Phosphatase Assays-- Phosphatase activity was determined by measuring the release of ³²P from radiolabeled casein or GST-R domain fusion protein (20). One unit of enzyme activity is defined as the amount catalyzing the release of 1.0 µmol of ³²PO₄ from radiolabeled GST-R domain/min at 30 °C. Briefly, substrates were phosphorylated by incubation with 0.3 µg of PKA in 50 mM Tris-HCl, pH 7.0, 10 µM magnesium acetate, 0.1% -mercaptoethanol, and 10 µCi of [-³²P]ATP for 16 h at 30 °C. The reaction was stopped by adding 0.1 ml of 100 mM EDTA and 100 mM sodium pyrophosphate, pH 7.0. To remove free [-³²P]ATP, the mixture was loaded on a Sephadex G-50 column that had been pre-equilibrated with 50 mM Tris-HCl, pH 7.0, 0.1 mM EGTA, and 5% glycerol. Phosphorylated substrate was collected by centrifuging the column (10,000 × g, 20 min). ³²P-labeled GST-R domain (3 µg) was incubated with a phosphatase (PP1, PP2A, PP2B, or PP2C, each at 20 nM) for 2 h, separated by SDS-PAGE on a 10% gel, and exposed to x-ray film. During assays, PP2B buffer contained 0.1 µM calmodulin and 0.8 mM CaCl₂; PP2C buffer contained 20 mM magnesium acetate.

To study the dephosphorylation of full-length CFTR, protein G-Sepharose beads (with bound antibody and CFTR) were preincubated with 1.5 µg of PKA, 20 µM ATP, 10 µg of bovine serum albumin, and 10 µCi of [-³²P]ATP and washed three times with 50 mM Tris solution. Radiolabeled CFTR was incubated with phosphatase (20 nM PP1, PP2A, PP2B, or PP2C) in appropriate assay conditions for 2 h, separated by SDS-PAGE on a 6.5% gel, and exposed to x-ray film.

Statistics-- Values are presented as the mean ± S.E. Significance was assessed at the 95% confidence level using the Student's t test.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Antibody to PP2C-- Four fusion proteins were prepared in an attempt to generate anti-PP2C antibody, but only one of these proved adequate for the studies described in this paper. Fig. 1A shows the purification of GST and of a GST-PP2C fusion protein (GST-PP2C₂₉₆) from BL21 cells. Fig. 1B is a Western blot probed with rabbit polyclonal antibody raised against GST-PP2C₂₉₆ (1:7, 500 dilution). The antibody recognized hamster (Fig. 1B, lane 2) and human PP2C (Fig. 1B, lanes 1 and 3) with similar affinity. Cross-reactivity was anticipated because of the close homology between human and hamster PP2C (21). The antibody also recognized purified turkey PP2C (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Purification of GST-PP2C296 and generation of anti-PP2C antibody. A, BL21 cells were transformed with plasmids directing expression of either GST or GST-PP2C296. Purification of these proteins was carried out on glutathione-Sepharose 4B as described under "Experimental Procedures." Samples (50 µg) from different steps of the purification procedure were resolved on 10% SDS-PAGE, and the bands were visualized by Coomassie Blue staining. Lane 1, total cell lysate from GST-expressing cells; lane 2, purified GST; lane 3, total cell lysate from GST-PP2C296-expressing cells; lane 4, purified GST-PP2C296; lane 5, products from thrombin cleavage of purified GST-PP2C296 (the PP2C peptide released by thrombin was not detectable). B, various PP2C-containing samples were resolved on 10% SDS-PAGE, transferred to nitrocellulose, and probed with antibody raised against purified GST-PP2C296. Lane 1, purified, recombinant human PP2C (20 ng); lanes 2 and 3, whole cell lysates (40 µg) from CFTR-transfected BHK cells and from human airway submucosal gland cell line Calu-3, respectively.

Distribution of Protein Phosphatases in the Membrane Fraction-- Those protein phosphatases that are tightly associated with CFTR should localize to the plasma membrane. To test for membrane localization of protein phosphatases, BHK cells were fractionated, and the membrane fraction was examined by Western blotting (Fig. 2, A-D). PP1, PP2A, PP2B, and PP2C were all detected in the membrane fraction. The level of PP2C in the membrane fraction relative to that in total cell lysates was somewhat lower than for PP1 and PP2B but much higher than PP2A, consistent with PP2A being predominantly a cytosolic enzyme (8). Similar results were obtained with Calu-3 and T84 cells. Control immunoblots using antibodies against marker enzymes did not reveal cytosolic or mitochondrial contamination in the membrane fraction from BHK cells. Na,K-ATPase (a plasma membrane marker) was highly enriched, whereas lactate dehydrogenase (cytosol) and cytochrome c oxidase (mitochondria) were detected in lysates but not in the membrane fraction (Fig. 2, E-G). Thus PP2C is present in membranes of BHK and other cells.

View larger version (49K): [in this window] [in a new window]   Fig. 2.   Subcellular distribution of phosphatases in BHK cells. BHK cell lysates (10 µg) and BHK membrane fractions (20 µg) were resolved on 10% SDS-PAGE, transferred to nitrocellulose, and probed with antibody specific for PP1 (A), PP2A (B), PP2B (C), or PP2C (D). Three other sets of blots were probed with antibodies specific for the following marker enzymes: E, Na+/K+-ATPase (plasma membrane); F, lactate dehydrogenase (LDH, cytosol); and G, cytochrome c oxidase (CCO, mitochondria). For comparison, purified phosphatases were loaded in the first lane of blots A-D as follows: A, rabbit PP1 (20 ng); B, turkey PP2A (40 ng); C, bovine PP2B (40 ng); and D, turkey PP2C (40 ng). The mass of each phosphatase is shown in parentheses.

Dephosphorylation of Full-length CFTR by PP2A and PP2C-- Immunoprecipitated CFTR was radiolabeled by incubation with PKA and [-³²P]ATP and then subjected to one of four phosphatases in the appropriate buffer (all at 20 nM): PP1 catalytic subunit (recombinant, bacterially expressed), PP2A1 or PP2B (bovine brain), or PP2C (turkey gizzard or bacterially expressed Paramecium PP2C). Samples were subjected to 6.5% SDS-PAGE, transferred to nitrocellulose membranes, and exposed to x-ray film. Although all phosphatases dephosphorylated CFTR (Fig. 3A), PP2A (lane 3) and PP2C (lanes 5 and 6) were noticeably more effective than PP1 (lane 2) and PP2B (lane 4), consistent with previous studies (9). Similar results were obtained using purified CFTR_His10 as the substrate (data not shown). Dephosphorylation followed the rank order: PP2C = PP2A > PP1 > PP2B, which is identical to the sequence observed for deactivation of CFTR channels by different phosphatases (11). Fig. 3B shows an immunoblot of the same membrane used in A for autoradiography, which indicates that comparable amounts of CFTR were loaded in each lane.

View larger version (82K): [in this window] [in a new window]   Fig. 3.   In vitro dephosphorylation of 32P-labeled CFTR by protein phosphatases. Immunoprecipitated CFTR was phosphorylated by a 16-h incubation with 1.5 µg of PKA and 10 µCi of [-32P]ATP and divided into 6 aliquots. Each aliquot was incubated for 1 h with control buffer (lane 6) or 20 nM protein phosphatase. Samples were separated by SDS-PAGE on 6.5% gels and transferred to nitrocellulose membrane. A, autoradiogram of the membrane showing residual 32P-labeled CFTR after treatment with PP1, PP2A, PP2B, turkey gizzard PP2C, or Paramecium PP2C. Lane 7 shows a control loaded with BHK total cell protein (40 µg, not radiolabeled). Positions of molecular mass markers are shown on the left. Arrows on the right indicate the position of fully glycosylated CFTR (band C). B, immunoblot of the same membrane shown in A probed with the monoclonal anti-CFTR antibody M3A7 to verify that similar amounts of CFTR were in aliquots exposed to phosphatases.

Dephosphorylation of Recombinant R Domain by PP2A and PP2C-- Recombinant R domain is potentially a more convenient substrate for phosphorylation/dephosphorylation studies, therefore we compared dephosphorylation of recombinant R domain by different phosphatases with that of full-length CFTR. A GST-R domain fusion protein (GST-R; predicted M_r = 52,000) was expressed in Escherichia coli BL21 cells and purified on a glutathione-Sepharose 4B column as described under "Experimental Procedures." Full-length GST-R protein was used for phosphorylation/dephosphorylation studies. To confirm that phosphorylation occurs mainly on the R-domain part of the fusion protein, equimolar GST-R and GST were exposed to 1.5 µg of PKA and 30 µCi of [-³²P]ATP, and the radiolabeling was compared by scintillation counting (Fig. 4A). Phosphorylation of GST was less than 3% of the fusion protein, suggesting that most phosphorylation of the fusion protein (i.e. >97%) occurs on the R domain fragment. GST-R was used as a phosphatase substrate in subsequent experiments. Fig. 4B shows that dephosphorylation of ³²P-labeled GST-R was most complete with PP2C (lanes 5 and 6) and PP2A (lane 3). PP1 also dephosphorylated GST-R (lane 2), whereas PP2B did not cause noticeable dephosphorylation (lane 4) at the same molar concentration (20 nM). Thus, sensitivity of the phosphorylated R domain to different protein phosphatases generally resembles that of intact CFTR (PP2C  PP2A  PP1 > PP2B), although GST-R may be less susceptible to PP2B.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Dephosphorylation of GST-R by different phosphatases. A, purified GST-R or GST was incubated with 1.5 µg of PKA and 30 µCi of [-32P]ATP. Phosphorylation of the two proteins was compared by scintillation counting. The data are plotted as cpm incorporated/mg protein (representative of two experiments). B, autoradiogram showing dephosphorylation of 32P-labeled GST-R domain fusion peptide (5 µg) by different phosphatases (20 nM). Lane 1 shows a phosphorylated fusion protein control (not exposed to phosphatase). Lanes 2-6 show aliquots from the same fusion protein preparation after exposure to PP1, PP2A, PP2B, PP2C, or Paramecium PP2C, respectively.

Co-immunoprecipitation of PP2C with CFTR-- The rapid deactivation of CFTR channels after excision hints that a phosphatase may be physically associated with the channel. To test this, CFTR was immunoprecipitated from BHK cells, and the immunoprecipitated proteins were analyzed by Western blotting with antibodies specific for each class of protein phosphatase. Untransfected BHK cells (lacking CFTR) served as negative controls. PP1, PP2A, and PP2B were not detected in immunoprecipitations of cells expressing CFTR (BHKwt·IP) or of control cells (BHK(-)·IP) but were easily observed in total cell lysates (BHKwt·t) (Fig. 5A-C). PP2C was also present in cell lysates (Fig. 5D; BHKwt·t), but unlike the other phosphatases, it was pulled down by the anti-CFTR antibody and was conspicuous in the immunoprecipitate (Fig. 5D, BHKwt·IP). This resulted from interaction with CFTR not antibody cross-reactivity, because PP2C was not co-precipitated when control cells lacking CFTR were used (Fig. 5D; BHK(-)·IP).

View larger version (53K): [in this window] [in a new window]   Fig. 5.   Co-precipitation of PP2C using anti-CFTR antibody. BHK cells were lysed and CFTR was immunoprecipitated using the anti-CFTR monoclonal antibody M3A7 as described under "Experimental Procedures." Immunoprecipitated proteins from CFTR expressing BHK cells (left lane) or from control untransfected cells (middle lane) were resolved on 10% SDS-PAGE, transferred to nitrocellulose, and probed with anti-PP1 (A), anti-PP2A (B), anti-PP2B (C), or anti-PP2C (D) antibody. The right lane of each blot contains 20 µg of total cell lysate.

The association between CFTR and PP2C was tested further by examining the ability of anti-PP2C antibody to co-precipitate CFTR from solubilized membrane extracts. CFTR was immunoprecipitated from the solubilized membrane fraction of BHK cells by anti-PP2C antibody (Fig. 6, lane 3). The low intensity of the CFTR band in lane 3 may indicate that only a small fraction of the immunoprecipitated PP2C is associated with CFTR. When solubilized membrane extracts from CFTR-expressing BHK cells were immunoprecipitated with the anti-PP2A antibody, CFTR was not co-precipitated (data not shown).

View larger version (48K): [in this window] [in a new window]   Fig. 6.   Co-precipitation of CFTR using anti-PP2C antibody. Membrane fractions were isolated from BHK cells, solubilized with Triton X-100, and immunoprecipitated using a polyclonal anti-PP2C antibody as described under "Experimental Procedures." Immunoprecipitated proteins were resolved on 6.5% SDS-PAGE, transferred to nitrocellulose, and probed with the anti-CFTR monoclonal antibody M3A7. Lanes 1 and 2, total lysate (10 µg) from CFTR-transfected and -untransfected BHK cells, respectively. Lanes 3 and 4, proteins immunoprecipitated from solubilized membrane fractions of CFTR-transfected and -untransfected BHK cells, respectively.

PP2C Can Be Cross-linked to CFTR_His10-- To further test if CFTR and PP2C are in close proximity, we examined whether they are cross-linked by the reagent DTSSP. DTSSP is a water soluble, thiol-cleavable cross-linker with a spacer arm length of 1.2 nm. Its two functional groups bond covalently with -amines, such as those in lysine side chains (22). The addition of DTSSP (400 µM) to lysates of BHK cells expressing CFTR_His10 resulted in the cross-linking of CFTR_His10 into high molecular mass complexes (Fig. 7A, lane 1). Some of the cross-linked CFTR_His10 could be released from these complexes by cleavage of the cross-links with DTT (Fig. 7A, lane 2). To identify proteins cross-linked to CFTR_His10, DTSSP-treated lysates were incubated with Ni²⁺-NTA-agarose and washed with buffer containing 40 mM imidazole. CFTR and associated proteins were eluted with 300 mM imidazole. Most CFTR appeared in the second elution fraction (E2, lane 5) and remained in high molecular mass complexes if the fraction was not treated with DTT prior to SDS-PAGE (E2A, lane 7).

View larger version (37K): [in this window] [in a new window]   Fig. 7.   Specific cross-linking of CFTRHis10 and PP2C by DTSSP. BHK cells expressing CFTRHis10 were lysed with 1% Triton X-100 and reacted with the cross-linker DTSSP. Ni2+-NTA-agarose beads were incubated batchwise with the cross-linked cell lysate, and the resin was loaded into a column. After washing, proteins were eluted with buffer containing 300 mM imidazole. Samples from different steps of the purification protocol were resolved on SDS-PAGE, transferred to nitrocellulose, and probed using anti-CFTR (A) or antiphosphatase (B) antibodies. In A, T and B indicate the top and bottom of the stacking gel, respectively. SM, starting material; FT, flow through. The positions of prestained molecular weight markers are indicated on the right. DTT was added to some samples to cleave DTSSP cross-links prior to electrophoresis so that CFTR and PP2C could be identified based on mass and immunoreactivity. In B, upper and lower immunoblots are from CFTRHis10-transfected and -untransfected BHK cells, respectively. The left lanes contain purified phosphatases as follows: rabbit PP1 (100 ng); bovine PP2A (1,000 ng); bovine PP2B (100 ng); and turkey PP2C (100 ng). Note that PP2C was the only phosphatase co-eluted with CFTR.

To determine whether phosphatases become cross-linked to CFTR_His10, immunoblots of the same fractions shown in Fig. 7A were probed with antibodies specific for PP1, PP2A, PP2B, and PP2C (Fig. 7B). DTT treatment of the samples prior to SDS-PAGE allowed proteins to be identified based on their apparent mass and reactivity. The upper blots in panel B show results obtained when lysate from BHK cells expressing CFTR_His10 (BHKwt) was cross-linked and Ni²⁺-NTA-purified. The lower blots show results obtained using control BHK(-) cells, which lack CFTR. Most phosphatase in the starting material was presumably not associated with CFTR_His10, because it was not retained by the Ni²⁺-NTA-agarose column and appeared in the flow-through. Of the four types of phosphatases examined by Western blotting (PP1, PP2A, PP2B, and PP2C), only PP2C was present in the eluate (Fig. 7B, top blot, lane E2). Importantly, PP2C was eluted in the second fraction, which was the same fraction containing most of the CFTR_His10 according to the blot in Fig. 7A. When the sample was not treated with DTT prior to electrophoresis, the PP2C band shifted from ~44 kDa to the junction between stacking and running gels (data not shown), where CFTR was also found (Fig. 7A). PP2C did not bind directly to the Ni²⁺-NTA-agarose because it was not detected in E2 if lysates were prepared from cells lacking CFTR (Fig. 7B, bottom blot). These results indicate that PP2C is specifically cross-linked to CFTR_His10 by DTSSP and can be co-purified by nickel chelate chromatography.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The present results demonstrate that PP2C is present in the membrane fraction of BHK cells and is closely associated with CFTR. We confirmed that phosphorylated CFTR is a good substrate for PP2C and found that the susceptibility of bacterially expressed GST-R domain fusion protein to different phosphatases is similar to that of full-length CFTR.

Rapid rundown of CFTR channel activity in excised membrane patches results from dephosphorylation of PKA sites; it can be reversed by exposure to the PKA catalytic subunit and does not occur if patches are excised into bath solution containing PKA (2). The association of CFTR with PP2C observed in the present work may be a widespread phenomenon because rundown with similar characteristics has been observed in many other cell types, including those from tissues affected in cystic fibrosis (e.g. pancreatic ducts (23), human colon (17), and human airway cells (3, 24).

Functional studies of excised patches indicate a major role for membrane-delimited PP2C but do not exclude regulation by PP2A and other phosphatases in situ. Deactivation of CFTR in excised patches is relatively insensitive to okadaic acid and calyculin A, independent of Ca²⁺ and calmodulin, and inhibited by lowering free magnesium concentration (2, 3, 11). Transepithelial studies are consistent with a predominant role of PP2C, because deactivation of the short-circuit current after removing forskolin or 8-(4-chlorophenylthio) adenosine 3':5'-cyclic monophosphate is unaffected by the PP1 and PP2A inhibitors okadaic acid (10) or calyculin A (11). Although co-expression of PP2C with CFTR in Fisher rat thyroid cells has little effect on the rate of chloride current deactivation after cAMP washout, it inhibits the cAMP-stimulated current by ~70% (10). The isoform of PP2C that regulates CFTR remains to be established. The antibody used in this study would be expected to recognize  and known splice variants of the  isoform because of their homology but not the more distantly related  isoform (25).

PP2A regulates CFTR in some cell types, particularly cardiac and sweat ducts (12, 26), and we confirmed that PP2C and PP2A are both more effective in dephosphorylating full-length CFTR and recombinant R domain peptide compared with equimolar PP1 or PP2B. However PP2C was more abundant in the membrane fraction from BHK cells, and no association between CFTR and PP2A was detected by immunoprecipitation or chemical cross-linking. By contrast, a specific monoclonal antibody against CFTR (M3A7 (27)) co-precipitated PP2C but did not bring down PP1 or PP2A. Polyclonal anti-PP2C antibody co-precipitated CFTR, whereas antibodies against PP1, PP2A, and PP2B did not. The association between PP2C and CFTR must be relatively strong, because immunoprecipitations were carried out in the presence of detergent (1% Triton X-100).

After exposure of cell lysates to the cross-linker DTSSP and purification by nickel chelate chromatography, CFTR_His10 appeared in high molecular mass complexes that also contained PP2C but not PP1, PP2A, or PP2B. Control experiments with untransfected BHK cells or cells expressing CFTR without a polyhistidine tail (data not shown) confirmed that PP2C binding to Ni²⁺-NTA-agarose requires CFTR_His10. The association between CFTR and PP2C does not require strong phosphorylation of CFTR, because all experiments were carried out using cells that had not been stimulated with cAMP.

A stable complex that includes CFTR and PP2C would presumably increase the efficiency of CFTR dephosphorylation. Such targeting of PP2C to particular substrates has not been reported previously, although other protein phosphatases form stable complexes with their substrates. For example, PP2B is associated with a neuronal potassium channel (28) and PP2A is targeted to the microtubule-associated protein tau 32 (29).

The monoclonal anti-CFTR antibody M3A7 used to co-precipitate PP2C in this study has been extensively characterized (27) and did not recognize proteins in control (untransfected) BHK lysates. The polyclonal antibody raised against PP2C was relatively specific but did recognize an unidentified 30-kDa protein, which may be another PP2C isoform.

Adding a polyhistidine tag to CFTR did not grossly alter its structure because CFTR_His10 had normal Cl channel activity when incorporated into planar bilayers.² Also, specific association of PP2C was observed with both wild-type and histidine-tagged CFTR. Any alterations in structure would seem more likely to disrupt than create specific interactions between the proteins, although that possibility cannot be formally excluded. CFTR and PP2C-like phosphatase activities were both present in the membrane fraction, and the Mg²⁺ dependence of channel rundown in excised patches provides further independent evidence that PP2C and CFTR are normally co-localized at the plasma membrane (11). Thus the selective co-precipitation and cross-linking were not artifacts of lysing the cells but likely reflect their co-localization under physiological conditions.

Close association of CFTR with PP2C could help explain the robust down-regulation of its channel activity in vivo when PKA stimulation is removed. The phosphatase-regulating CFTR has been suggested as a target for pharmacotherapies aimed at increasing (i.e. restoring) CFTR activity in cystic fibrosis, therefore these observations may have relevance to the development of therapies for cystic fibrosis (3, 30, 31). A phosphatase inhibitor might be useful for stimulating mutant channels that reach the plasma membrane when used alone or as an adjunct therapy to increase the efficacy of other treatments such as drugs that improve CFTR processing or gene therapies. Regardless, ion channel rundown is a common phenomenon in patches excised from neuronal, cardiac, and other cell types. It will be interesting to learn if PP2C also forms regulatory complexes with other ion channels besides CFTR and mediates their deactivation.

	FOOTNOTES

^* This work was supported by the NIDDK, National Institutes of Health Grant DK54075-03.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Postdoctoral fellow of the Canadian Cystic Fibrosis Foundation (CCFF).

^§ Contributed equally to this work.

^¶ Recipient of a CCFF graduate studentship.

Senior scientist of the Medical Research Council (Canada).

^** To whom correspondence should be addressed: Dept. of Physiology, McGill University, 3655 Drummond St., Montréal, Québec H3G 1Y6, Canada. Tel.: (514) 398-8320; Fax: (514) 398-7452; E-mail: hanrahan@med.mcgill.ca.

² D. Dahan, unpublished observation.

	ABBREVIATIONS

The abbreviations used are: CFTR, cystic fibrosis transmembrane conductance regulator; PKA, cAMP-dependent protein kinase; PP1, type 1 phosphatase; PP2, type 2 phosphatase; BHK, baby hamster kidney; NTA, nitrilotriacetic acid; GST, glutathione S-transferase; DTSSP, dithiobis(sulfosuccinimidyl propionate); PAGE, polyacrylamide gel electrophoresis; R, regulatory; DTT, dithiothreitol; E1, E2, E3, first, second, and third elution fractions; wt, wild type.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				L. A. Borthwick, J. Mcgaw, G. Conner, C. J. Taylor, V. Gerke, A. Mehta, L. Robson, and R. Muimo The Formation of the cAMP/Protein Kinase A-dependent Annexin 2 S100A10 Complex with Cystic Fibrosis Conductance Regulator Protein (CFTR) Regulates CFTR Channel Function Mol. Biol. Cell, September 1, 2007; 18(9): 3388 - 3397. [Abstract] [Full Text] [PDF]

				P. M. Haggie, J. K. Kim, G. L. Lukacs, and A. S. Verkman Tracking of Quantum Dot-labeled CFTR Shows Near Immobilization by C-Terminal PDZ Interactions Mol. Biol. Cell, December 1, 2006; 17(12): 4937 - 4945. [Abstract] [Full Text] [PDF]

				I. R. Bates, B. Hebert, Y. Luo, J. Liao, A. I. Bachir, D. L. Kolin, P. W. Wiseman, and J. W. Hanrahan Membrane Lateral Diffusion and Capture of CFTR within Transient Confinement Zones Biophys. J., August 1, 2006; 91(3): 1046 - 1058. [Abstract] [Full Text] [PDF]

				W. R. Thelin, M. Kesimer, R. Tarran, S. M. Kreda, B. R. Grubb, J. K. Sheehan, M. J. Stutts, and S. L. Milgram The Cystic Fibrosis Transmembrane Conductance Regulator Is Regulated by a Direct Interaction with the Protein Phosphatase 2A J. Biol. Chem., December 16, 2005; 280(50): 41512 - 41520. [Abstract] [Full Text] [PDF]

				S. Liu, A. Veilleux, L. Zhang, A. Young, E. Kwok, F. Laliberte, C. Chung, M. R. Tota, D. Dube, R. W. Friesen, et al. Dynamic Activation of Cystic Fibrosis Transmembrane Conductance Regulator by Type 3 and Type 4D Phosphodiesterase Inhibitors J. Pharmacol. Exp. Ther., August 1, 2005; 314(2): 846 - 854. [Abstract] [Full Text] [PDF]

				S. G. Bompadre, T. Ai, J. H. Cho, X. Wang, Y. Sohma, M. Li, and T.-C. Hwang CFTR Gating I: Characterization of the ATP-dependent Gating of a Phosphorylation-independent CFTR Channel ({Delta}R-CFTR) J. Gen. Physiol., March 28, 2005; 125(4): 361 - 375. [Abstract] [Full Text] [PDF]

				N. C. Connors, M. E. Adams, S. C. Froehner, and P. Kofuji The Potassium Channel Kir4.1 Associates with the Dystrophin-Glycoprotein Complex via {alpha}-Syntrophin in Glia J. Biol. Chem., July 2, 2004; 279(27): 28387 - 28392. [Abstract] [Full Text] [PDF]

				C. Li, K. Roy, K. Dandridge, and A. P. Naren Molecular Assembly of Cystic Fibrosis Transmembrane Conductance Regulator in Plasma Membrane J. Biol. Chem., June 4, 2004; 279(23): 24673 - 24684. [Abstract] [Full Text] [PDF]

				M. Flajolet, S. Rakhilin, H. Wang, N. Starkova, N. Nuangchamnong, A. C. Nairn, and P. Greengard Protein phosphatase 2C binds selectively to and dephosphorylates metabotropic glutamate receptor 3 PNAS, December 23, 2003; 100(26): 16006 - 16011. [Abstract] [Full Text] [PDF]

				P. Ofek, D. Ben-Meir, Z. Kariv-Inbal, M. Oren, and S. Lavi Cell Cycle Regulation and p53 Activation by Protein Phosphatase 2Calpha J. Biol. Chem., April 11, 2003; 278(16): 14299 - 14305. [Abstract] [Full Text] [PDF]

R. Ganeshan, A. Di, D. J. Nelson, M. W. Quick, and K. L. Kirk The Interaction between Syntaxin 1A and Cystic Fibrosis Transmembrane Conductance Regulator Cl- Channels Is Mechanistically Distinct from Syntaxin 1A-SNARE Interactions J. Biol. Chem., January 24, 2003; 278(5): 2876 - 2885. [Abstract] [Full Text] [PDF]